Financial Risks, Stability, and Globalization


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

The free flow of capital is a reality in the era of globalization. It follows naturally from the international integration that increasing trade flows and technological advancements in communications have brought about. In this new environment, the challenge public authorities face is not how to restrict international capital flows, but how to prepare domestic conditions so that capital flows benefit the country the most or, alternatively, hurt it the least. The specific ways in which a policymaker should address the issues posed by capital flows depend on the particular area of the policymaker’s responsibility, whether banking supervision and regulation, fiscal and tax policies, or monetary management. The paper by Sundararajan, Ariyoshi, and Ötker-Robe provides a well-argued, complete, and comprehensive review of policy measures aimed to reduce vulnerabilities (and, on the contrary, profit) from the era of open capital accounts. I congratulate the authors for such a fine paper. However, if the strength of the paper lies in being comprehensive, that is also its main weakness. For actual policymaking, it is necessary to make precise and operational the concepts of good management proposed in the paper.

Take, for example, the general recommendation of keeping “a consistent macroeconomic policy mix” set forth by the authors. My claim—and this is the first part of my comment—is that, if we are worried about potential unstability brought about by external financial shocks in a fully integrated economy, we should be concerned with which “consistent macroeconomic policy mix” minimizes such unstability. From my perspective as a central banker, I would like to discuss some aspects of monetary policies to achieve domestic price stability, as part of a consistent macroeconomic policy mix, and how this relates to the economy’s degree of openness. In the second part of my comment, I will address some topics concerning controls on the capital account.

Inflation Targeting and External Shocks

Since ten years ago, the Central Bank of Chile has followed a monetary regime based on inflation targeting. In fact, Chile is the second country in the world (after New Zealand) to have implemented such a regime. Recent monetary policy in Chile has become even more committed to inflation targeting by making the exchange rate fully flexible (through the elimination of exchange rate bands and central parities in September 1999) and by publishing its first, formal “inflation report” (Informe de Política Monetaria, May 2000). Many countries around the world have become convinced of the merits of inflation targeting and have adopted it in recent years. Examples are Canada (1991), the United Kingdom (1992), Israel (1992), Spain (1994), and, in the past two years, the Czech Republic (1998), Hungary (1998), Poland (1999), and Brazil (1999).

The question I would like to discuss is to what extent inflation targeting is consistent or can be made consistent with reducing vulnerabilities in a context of open capital accounts. The initial theoretical arguments in favor of inflation targeting have been based on models that assume closed economies. For example, Svensson (1997) and Ball (1997) present models where inflation targets implemented through Taylor rules (i.e., rules by which interest rates are adjusted in response to output and inflation) are optimal policies. Only recently have researchers addressed the merits of inflation targeting in an open economy context. In a very influential paper, Larry Ball (1998) argues that in economies subject to external shocks, rigid inflation targets can lead to higher output and exchange rate volatility. The mechanisms can be outlined as follows. Consider an external shock that moves the exchange rate. This, in turn, has an effect on import prices and, thus, consumer price index (CPI) inflation (this is the well-known “pass-through” effect). If the monetary authority has rigid, short-horizon inflation targets, it will react strongly to the price changes induced by the external shock. This reaction will both induce a large swing in exchange rates (given that sizable shifts in import prices are needed to move the average price level) and exacerbate the volatility of output and real activity in general (through various interest-rate-related mechanisms). Thus, the zeal to stabilize domestic prices would produce an amplification of the effects of the original external shock. Is there empirical evidence that supports these theoretical arguments? In last year’s international conference of the Central Bank of Chile, Cecchetti and Ehrmann (2000) presented an empirical cross-country paper that concluded that inflation-targeting countries had been more successful than others in both bringing inflation down and keeping inflation volatility low. However, they also found that inflation-targeting countries experienced larger output volatility, thus describing a trade-off between lower inflation volatility and wider output fluctuations.

Should we conclude that inflation targeting is not a good monetary regime for open economies? Not necessarily. Inflation targeting has been instrumental to achieving price stability, a long-desired goal, in many countries. The question is how to modify it so that it delivers the objective of price stability without compromising the general well-being of the economy in a context of open trade and capital markets. Some research has started to tackle this difficult question and promises to help significantly the design of monetary policy in the era of globalization. This research should help to answer questions such as, What is the optimal timing and intensity of monetary response to domestic and external shocks? Should inflation targeting during the transition to low inflation be conducted differently from when prices are stable? What are the policy coordination issues we should keep in mind when implementing inflation targeting?

After reviewing this theoretical and empirical work, I would like to emphasize that many key questions for a central banker working in an open economy remain unanswered, and more research is badly needed. Let me give you an example. Some authors appear to recommend “flexible” inflation targeting. But before we accept their recommendations, we must be aware that both work in models in which the economy has already achieved what Svensson calls an “era of price stability.” In technical terms, their theories focus on “steady states.” For countries still facing problems of high and volatile inflation, is flexible inflation targeting still the best regime? And is the source of shocks (domestic or external) the main determinant of the type of inflation targeting to be followed? Other authors (see, for instance, Kumhoff, 1999) argue that the key issue to determining the best monetary regime is whether the central bank has achieved credibility in pursuing price stability.

I have undertaken this long discussion of inflation targeting in a context of open trade and capital accounts to highlight the point that it is necessary, but not enough, to say that countries need “a consistent macroeconomic policy mix” (or, by extension, that a “strong banking supervision” should be in place). Where the main difficulty for policymakers resides and where we need stronger research and advice is on the subtleties and details of policy implementation.

Regulations to the Capital Account

Chile is well known for its policy of regulating the capital account through a nonremunerated reserve requirement—NRRR—on (short-term) capital inflows. After eight years of applying it at a rate of 30 percent, the measure was suspended in 1998, in the middle of the severe global financial turmoil that marked the period. Many papers have been written on the efficiency and efficacy of the NRRR in Chile, the main conclusion being that it was not, with time, effective in impeding massive capital inflows and a real exchange rate appreciation (its original purpose), although it did contribute to bias those inflows toward the medium to long term and away from short-term capital (its secondary objective). What I want to stress in this respect and concerning the paper by Sundararajan, Ariyoshi, and Ötker-Robe are two points: first, sequencing; and second, the consistency of the NRRR with the macroeconomic policy mix.

With respect to sequencing, the paper advises advancing gradually toward full integration of the capital account while putting in place all the remaining ingredients to make local financial markets resilient to external volatility. Although this logic was somewhat in place during the 1990s in Chile, in practice it posed a very clear trade-off between keeping the pace of gradual capital account opening and the efficacy of the NRRR. In effect, the gradualist path implied that other restrictions were reduced or eliminated as time went by, facilitating capital inflows and outflows, thus helping the market to find ways to circumvent the NRRR. And whenever there was a doubt as to whether to extend the NRRR coverage or not, the inclination was more toward not doing so, because that would have implied a reversal in the path to full integration. Therefore, any transition of this sort should be made as brief as possible; otherwise the alleged “protection” brought by an instrument like the NRRR will fade away before one can notice.

In the same vein, the NRRR, by visibly punishing more short-term inflows in the form of bank credit, tends to enhance the disintermedi-ation of the local financial sector. Indeed, in the case of Chile and until around 1998, domestic corporations were increasingly resorting to financing their investment needs by directly placing equity (not subject to NRRR) and bonds (long-term nature) in world financial centers, rather than looking for foreign loans (less long-term) channeled through the local financial system. Although this fact was a blessing when the Asian crisis erupted, it did not do much to help the development and progress of domestic financial markets, which are clearly objectives of gradually opening capital accounts.

And second, as already indicated, the NRRR was originally placed as an instrument to keep the peso from appreciating in real terms, when large capital resources started to flow in, based on the conviction that a depreciated peso was essential to export and overall growth. The macro policy mix then was also attempting to gradually reduce inflation from moderately high levels to single-digit levels, which required keeping rather high interest rates on average, which in turn tended to attract more capital from abroad and strengthen the peso. In the meantime, a crawling exchange rate band tried to signal where the central bank wanted the peso to move along (and the fiscal stance was neutral with respect to the cycle and in surplus overall). In sum, trying to keep two objectives simultaneously (a declining inflation and a certain diffuse path for the real exchange rate) asked for more than one powerful instrument (the pure monetary policy), and the massive nature of inflows clearly overwhelmed the exchange rate band. So that was the main logic for implementing the NRRR early in the 1990s. The point I would like to make is that if one can dispose of the real exchange rate path as a central bank objective, allowing it to concentrate on stabilizing prices in the medium term, then free exchange rate fluctuations can do the job of preventing short-term volatile capital inflows. This is so provided that the country’s authorities facilitate the private sector design and implementation of exchange rate risk-hedging mechanisms (including the placement of local-currency-denominated foreign debt). If the size and sophistication of local financial markets will allow this at an early stage of the process remains to be seen; as the authors rightly state, the more you delay this kind of liberalized option, the less likely it is that the local financial markets will ever be able to develop.

Let me close by repeating my congratulations to the authors for this very thorough paper.


    BallLaurence1997Efficient Rules for Monetary Policy,NBER Working Paper No. 5952 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

    BallLaurence1998Policy Rules for Open Economies,NBER Working Paper No. 6760 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

    CechettiStephen and MichaelEhrmann2000Does Inflation Targeting Increase Output Volatility? An International Comparison of Policymakers’ Preferences and Outcomes,Central Bank of Chile Working Paper No. 69 (Santiago: Central Bank of Chile).

    • Search Google Scholar
    • Export Citation

    KumhoffMichael1999Inflation Targeting under Imperfect Credibility” (unpublished; Palo Alto, California: Stanford University).

    SvenssonLars E.O.1997Inflation Forecast Targeting: Implementing and Monitoring Inflation Targets,European Economic Review Vol. 41 (June) pp. 111146.

    • Search Google Scholar
    • Export Citation

    Other Resources Citing This Publication