II.Capital Controls on Inflows

　　This section begins with a brief discussion on whether capital account liberalization is really beneficial to an economy. It then discusses about sequencing issues. The remaining section focuses on the capital controls adopted in Chile and examines whether such controls were effective in reducing short-term and total capital inflows.
　　

1.Various Views on Capital Account Liberalization
　Effects of Capital Account Liberalization

　　Various scholars have pointed out a number of benefits from capital account liberalization. First, international financial transactions transfer resources from high-saving to low-saving countries. The transfer helps smooth the time profile of consumption and investment under business-cycle disturbances and allow firms and households to diversify country-specific risks (Eichengreen, 1999a). Obstfeld and Rogoff (1996) also argued that gains from intertemporal trade can be increased by developing a free international market for securities. This is because volatility of domestic output caused by various unexpected shocks could be reduced by gaining access to securities, thereby stabilizing consumption patterns and current account levels.

　　Second, Klein and Olivei (1999) emphasized that open capital markets enhance welfare and increase efficiency through better resource allocation.

　　Third, foreign borrowing and lending activities contribute to financial sector development through increasing the volume of transactions and thus lowering transaction costs (Klein and Olivei, 1999). A drop in transaction costs encourages economic agents to undertake lending and borrowing activities, thereby developing financial markets. A well functioning financial system mitigates the problems of asymmetric information and lowers transaction costs. This in turn enables an economy to mobilize domestic savings, to realize an efficient allocation of resources, to develop risk management skills, to strengthen corporate control, and to facilitate trade of goods and services.

　　Other scholars, however, stress several disadvantages. First, Bhagwati (1998) and Cooper (1998) argued that in a world with imperfect information and imperfect financial markets, free capital mobility is likely to amplify existing distortions arising from information asymmetry. This takes place through increasing the opportunities for herd behavior as well as the scale of such a phenomenon. Also, capital account liberalization encourages excessive risk taking activities, increasing the likelihood of crises.

　　Second, Eichengreen and Wyplosz (1996) and Frankel (1999) pointed out that most foreign exchange transactions have little to do with economic fundamentals, and only contribute to destabilizing and reducing social welfare. For this reason, they make a case for the Tobin tax as a tool to lower welfare-reducing short-term capital flows without affecting welfare-enhancing long-term flows***. The Tobin tax is likely to reduce overall foreign exchange transactions regardless of whether the transactions are associated with inflows into or outflows from particular countries. To be effective, however, the Tobin tax has to be implemented by all countries simultaneously. This suggests that the Tobin tax is politically and technically impractical. Therefore, the Tobin tax does not address the issue of massive capital inflows into individual emerging-market economies.

　　Third, according to Rodrik (1998), no empirical evidence supports the view that an economy that liberalized the capital account─by removing restrictions on payments for capital transactions─grew faster. Meanwhile, it is self-evident that international financial liberalization exposed countries to the danger of a debilitating crisis. While this result constitutes some evidence against capital account liberalization, Eichengreen (1998) argues that the failure to find significant positive results of capital account liberalization may be attributed to the possibility that some variables that were negatively associated with growth but positively associated with capital account liberalization were omitted.

　　Fourth, Klein and Olivei (1999) showed that in the cases of Latin American countries, there was no significant effect of capital account liberalization on financial deepness during 1986-1995. This result is contrasted with OECD countries, where capital account liberalization positively affected the deepening of the financial sector and economic growth during the same period. This suggests that a constellation of economic, administrative, and legal institutions in needed in order to promote financial deepening and generate a positive impact of capital account liberalization on economic growth.

Capital Controls: Rationale and Effects

　　Given the lack of strong empirical support for capital account liberalization, there are at least three reasons in support of capital controls (Nadal-De Simone and Sorsa, 1999). First, capital controls can be used as a way to gain monetary policy independence. However, the efficacy of monetary policy in creating a wedge between domestic and foreign interest rates is reduced particularly when an asset market adjusts promptly and high substitutability exists between domestic- and foreign-currency denominated assets.

　　Second, capital controls can be justified as a tool to improve social welfare For example, volatile capital movements amplify the problems of asymmetric information, which often gives rise to distortions in competitive equilibrium by making private investors prone to noise trading, inducing herd behavior, and amplifying the boom-and-bust cycle. Implicit government guarantees of banks' external liabilities under asymmetric information between banks and stockholders (and depositors) also generate welfare-reducing results.

　　Third, Johnston and Otker-Robe (1999a), Eichengreen, Mussa et al. (1999), and Eichengreen (1998, 1999) stressed that capital controls can be used as a third line of defense following the first line of defense (bank's own risk-management practices) and the second line of defense (regulatory supervision). Financial liberalization is inevitable for a country that wishes to take advantage of benefits from higher investment, faster growth, and better living standards. The problem, however, lies in inefficient financial markets that result from moral hazard, adverse selection, and herd behavior─all of which are likely to be amplified by volatile capital movements. These problems can be mitigated if proper policies─including taxes, bank supervision, regulation and appropriate government guarantees─are implemented.

　　However, it takes time to properly design and successfully implement such policies. For example, it took about ten years for Chile to develop a sound banking sector with adequate prudential supervision and regulations. Therefore, Eichengreen (1998, 1999a) claimed that capital controls are justified on prudential grounds and justifiable only where financial markets are thin, private sector's risk management practices are underdeveloped, and regulators' capacity to supervise the financial sector is limited. To deepen financial markets, banks should acquire sophisticated risk-management skills and regulators should gain experience, competence and independence. Once these measures are adequately implemented, capital controls should be removed. Thus, it is important to stress that capital controls should not be used as an excuse to slow down institutional development and policy reform.

Sequencing Issues

　McKinnon (1973) argued that opening of the capital account through dismantling capital controls should be postponed until free trade in goods is consolidated. If the opposite sequencing takes place, massive capital inflows caused by capital account liberalization result in a real exchange rate appreciation, which would frustrate a move toward opening up trade in goods. McKinnon and Pill (1996) emphasized that capital account liberalization should be postponed until the end of the banking sector reform process, because of moral hazard issues arising from explicit or implicit deposit protection.

　　Eichengreen, Mussa, et al. (1999) claimed that capital account liberalization per se is not responsible for an increase in costly economic crises. The problem, in fact, is associated with an inappropriate sequencing of capital account liberalization; that is, the removal of capital controls before strengthening the domestic financial system. Compared with advanced industrial countries that have opened their capital accounts and developed a deep, mature and efficient domestic financial system, the experiences of developing countries indicate that it is crucial to liberalize the capital
account only after strengthening the domestic financial sector. Calvo (1998a) argued that poorly regulated banks intermediate inflows of capital in an inefficient or corrupt way, thereby increasing the probability of a systemic financial crisis.

　　Radelet and Sachs (1998) expressed the view that a rapid push toward fully open capital markets is misguided if it allows short-term capital flows among developing countries. This is because there is no strong empirical evidence in support of the claim that economic growth in middle-income developing countries depends on unfettered access to short-term capital flows from abroad. Radelet and Sachs argue that short-term financing is useful for trade flows but not for longer-term investments.

　　Cooper (1998) suggested that capital controls should be used to structure maturities by promoting direct equity investment (FDI) first, followed by accepting long-term bonds but avoiding short-term borrowing. Similarly, Sachs and Woo (1999) suggested that capital controls on short-term inflows are advisable while long-term capital inflows should be promoted. In addition to FDI-related inflows, there is solid empirical evidence that inflows of equity portfolio capital have positive effects on investment and economic growth (Williamson, 2000). An expectation of pending liberalization causes a stock market boom that reduces the cost of equity capital, while the lower cost of equity capital following actual liberalization is typically succeeded by an important though temporary (three-year) investment boom (Henry, 1999a and 1999b). A swap of portfolio equity capital can be strongly welfare-enhancing by allowing both parties to diversify risk. FDI outflows can also allow national firms to exploit their company-specific assets, such as patents, technology, and management know-how on the world stage (Williamson, 2000).

2.The Chilean Case

　　 In 1991 Chile introduced capital controls to contain massive capital inflows particularly through adopting unremunerated reserve requirements and minimum holding periods on short-term borrowing from non-residents. The controls aimed at achieving four main purposes. They were: (1) slow down the volume of capital inflows and tilt its composition towards longer maturities; (2) reduce or at least delay any real exchange rate appreciation stemming from these inflows; (3) allow the central bank to maintain a large differential between domestic and international interest rates and thus to conduct an independent monetary policy; and (4) reduce the vulnerability to international financial instability (Edwards, 1999a).

Economic Reforms since mid-1980s

　Chile recovered from the 1981-82 crisis after the mid-1980s and was gradually able to relax the foreign exchange constraint imposed during the crisis (Marfan and Bosworth, 1994). Export growth was promoted by the real depreciation that began in 1982. The debt burden was reduced by the agreements on the postponement of payments and debt rescues. After 1985, the terms of trade improved and international interest rates dropped. In December 1989, a new government was democratically elected after seventeen years of military rule.

　　Given the favorable internal and external developments, the government introduced a comprehensive adjustment program in the mid-1980s. The purposes of the program were to avoid prolonged credit and consumption booms, to mitigate sharp appreciations and misalignments of exchange rates, to prevent excessive current account deficits, and to contain domestic and external debt.

　　The economic policies adopted under the program were as follows. First, the government undertook comprehensive banking sector reform. The government introduced a banking regulatory and supervisory scheme through the enactment of a new banking law in 1986. Under the law, strict guidelines on banks' exposure and activities with on-site inspections were adopted. The new banking law was amended in 1989. A securities law was introduced in 1993 with the purpose of increasing transparency in capital markets and regulating conflicts of interest. In 1997, a new law was introduced to widen banks' activities and to set rules for the internationalization of the banking system. These legal developments helped to foster sound development of the banking sector and to increase credit from the banking to the private sector, which rose after 1985 when the economy recovered. As a result, nonperforming loans dropped from 9.65% in 1984 to 1% in 1996. Also, tight prudential supervision and regulations of financial institutions contributed to the deepening of the capital market and enabled the country to borrow at rates closer to those of developed countries (Perry and Leipziger, 1999).

　　Second, since the mid-1980s the central bank adopted a real interest rate targeting approach to contain inflation. The peso-denominated short-term interest rate was maintained above a foreign interest rate. Short-term interest rates were imperfectly indexed to the monthly change in the Consumer Price Index through daily adjustments in an artificial unit of account known as the Unidad de Fomento (UF) in the 1980s. The central bank achieved independence in 1989. In 1990, the government undertook the so-called "sobreajuste (over-adjustment)" policy by raising the indexed interest rate from 8% to 16%. Despite the tight monetary policy, inflation increased for several months due to the fact that any small surge in inflation was rapidly magnified by the extremely high degree of indexation. All prices of durables were quoted in UF and some services like school fees were also quoted in UF.

　　In the late 1980s, the central bank aimed at influencing short-term interest rate as anchor interest rates so as to produce desired targets of monetary aggregates (Calvo and Mendoza, 1998). In the early 1990s, the central bank switched to a system of direct sales of 90-day bills that constituted liabilities of the central bank. From 1995, the anchor short-term rate was changed to the overnight interbank interest rate. Once this key rate was set, the term structure on government obligations was market-determined in auctions of longer-term papers. The target level of the anchor interest rate was set to a level that is consistent with objectives for growth and inflation, and from 1995 it was set to a publicly announced inflation target. In 1996, the central bank switched from an interest rate targeting approach to an inflation targeting approach by influencing an overnight indexed interbank interest rate to conform with the preannounced annual inflation target. Recently, the central bank tried to gradually shift its emphasis to nonindexed instruments. This policy is complemented with close monitoring of narrow monetary aggregates.

　　Third, the central bank introduced a new crawling band exchange rate regime─maintaining the peso within a band─in 1985. The main purpose of this exchange rate policy was to maintain the international competitiveness of Chilean exports. The central parity was indexed to the productivity-adjusted monthly differential between Chile's CPI inflation and that of its largest trading partners through daily adjustments. Although the fluctuation was within a band, the band was progressively widened and shifted if the central bank was convinced that the equilibrium exchange rate was outside the band. The band was initially set at ±2%, shifting to±3% in 1988 and further to ±5% in 1989. The gradual strengthening of the peso gave exporters some time to adjust and improve their competitiveness, realizing high average productivity and export growth. The equilibrium rate was defined as the rate achieving full employment that would induce a current account deficit of 3-4% of GDP.

　　The central bank maintained the pragmatic approach described above even when massive capital inflows took place from the late 1980s. The use of the exchange rate as an anchor was abandoned in 1996 in an effort to bring inflation down at any cost. After the Mexican crisis and the Asian crisis, the peso was allowed to fall with very limited intervention and monetary tightening. The change in the policy aimed at avoiding any implicit exchange rate insurance to dampen speculative pressures arising from interest rate differentials (Eyzaguirre and Lefort, 1999). Chile eventually adopted a floating regime in 1999.

　　Fourth, the government attempted to keep a small surplus in the trend level of the overall fiscal account after netting out cyclical fluctuations, such as copper price. The government established the Copper Stabilization Fund in order to save a faction of the windfall gains expected when the copper price was above a certain bench mark (i.e., long-term price). With the government's strong commitment, this fund helped mitigate instability arisen from the volatility of the terms of trade. Until 1997, the fund reached about 2% of GDP. The fiscal surplus was achieved from 1989 to 1998 and was sufficient to buy back the full foreign debt of the government and central bank.

　　Despite the fiscal reforms, some problems remain. For example, central bank incurred losses as a result of the low-yield assets acquired as part of rescuing the banking system in the mid-1980s and partial sterilization of large capital inflows in the 1990s. Furthermore, the Ministry of Defense collects 1/10 of the revenue of the government's copper company without any accountability on its use.

　　Fifth, the government continued to make progress on trade reform in the past two decades, with the exception of when import tariff rates were raised in the crisis aftermath of 1981-82. However, the government restarted trade reform in 1984 and established a uniform rate of 15% by 1987. The increased competition as a result of trade liberalization and a real depreciation led to export growth. In 1991, the government reduced import tariffs further to a uniform 11% and firmly pushed for the creation of a free trade zone with the United States (Dornbusch and Edwards, 1994). Trade reforms after 1984 were supported by a strongly depreciated real exchange rate. The real exchange rate was devalued by about 80% in 1982-88.

　　Sixth, the government developed capital market from the mid-1980s. Bank deposits rose from 30% of GDP in the mid-1980s to about 45% in 1996. Over the same period, government bonds as a share of GDP rose from 13% to 33%, private bonds increased from 7% to 15%, and paid stocks grew from 13% to 95% (Eyzaguirrre and Lefort, 1999). Reflecting the maturity of the capital market, the economy had a relatively low bank inter-mediation ratio. The successful financial market was the outcome of sound macroeconomic policies, stringent banking and securities regulations and supervision, and, particularly, pension reform. In 1981, the government shifted from the old pay-as-you-go-system to a funded scheme. Under the reform, the pensions of the generation who had already retired were mainly financed by fiscal revenue, while the contributions of active employees were accumulated as fresh savings in the private pension funds. Furthermore, the deficit from the old system was fully paid by the budget, amounting to yearly expenses of some 3% of GDP. Private pension funds amounted to about 40% of GDP in 1996, contributing to the rapid development of the domestic capital market.

Capital Controls Since 1991

　　Thanks to the successful economic program described above, Chile achieved economic growth of 10% in 1989, from 2.4% in 1985. The rate of unemployment also dropped from 12% in 1985 to 5.3% in 1989. Reflecting these favorable conditions, Chile faced massive capital inflows from the late 1980s. The surge in capital inflows was compounded by a general wave of international capital flowing to emerging economies. However, substantial capital inflows made the central bank difficult to maintain the aforementioned real interest rate targeting. In response, Chile reintroduced restrictions on inflows in 1991 in order to maintain a gap between a domestic interest rate and a foreign interest rate.

　　First of all, a 20% reserve requirement was imposed on new foreign borrowing and applied to foreign loans to banks and nonbanks. Trade credit was exempted from the requirement provided that shipment took place within six months. A holding period was also introduced and set equal to the loan maturity with a minimum of three months and a maximum of one year.

　　In July 1992, the reserve requirement was raised to 30% and the minimum holding period was set at one year. Also, the coverage was extended to trade credit and loans related to FDI projects. In 1995, capital controls were strengthened further by extending the coverage to the Chilean stocks traded in the New York Stock Exchange Market and to international bond issues. In June 1998, the reserve requirement was lowered to 10% after Chile began to face capital outflows. In September 1998, the reserve requirement was further lowered to 0%. With respect to FDI inflows, Chile imposed a three years minimum stay period to one year. While FDI-related loans were subject to capital controls, no restrictions were imposed on the repatriation of profits from FDI projects.

　　Edwards (1999a) estimated the tax equivalent for foreign funds that stayed in Chile based on the reserve requirements, the length of time to stay, and the opportunity cost. He found that the shorter the length of time that investment funds remain in the country, the higher the implicit tax rate was. Also, he stressed that the implicit tax rate rose sharply in 1994-95 and was high even for longer maturities. For example, the average implicit tax was 80 basis points even for three-year investments in 1997.

　　Thus, the initial coverage of the reserve requirement was partial. The initial protection of trade credits and other non-debt creating capital flows reflected strong support for export development by the Constitution or for practical reasons (e.g., the difficulty to control informal flows). Also, various types of foreign borrowing were exempted from capital controls initially. For example, inflows of less than US$10,000 were exempted for the purpose of facilitating trade credit (Eichengreen, 1999a). This invited investors to break up larger transactions into smaller ones for the purpose of evasion.

　　Therefore, the government extended the coverage to virtually all forms of foreign financing other than FDI in 1995. Foreign banks then established domestic investment firms to exploit the FDI loophole. In response, the government stopped authorizing financial FDI-related projects after mid-1995. These events indicate that capital controls should minimize exemptions and that the controls are likely to become ineffective in the end due to circumvention.

*Dean,Asian Development Bank Institute,Tokyo,Japan.
**Visiting Scholar,Asian Development Bank Institute and Associate Professor of Economics,Keio University, Tokyo,Japan.
***The Tobin tax refers to a global tax on foreigh exchange transactions. Tobin's proposal(1978)aims at reducing destabilizing speculation in international financial markets. This idea was also extensively discussed in the context of the devaluation of the pound sterling and the Italian Lira in the mid-1990s.Eichengreen and Wyplosz
(1993)state that the Tobin tax would discourage short-term-oriented speculators from betting against major currencies.Brent(1999)stresses that the Tobin tax rate should be very high to be effective.

下載全文預覽列印