Printer Friendly

Black market exchange rates and capital mobility in Asian economies.


Since the early 1980s, many developing countries in the Asian Pacific region have been under pressure to open their domestic financial markets and dismantle their capital controls. Singapore and Malaysia abolished foreign exchange controls in the late 1970s. Japan embarked on a liberalization scheme in 1981, followed by Indonesia and the Philippines in the early 1980s. Thailand, Korea, and Taiwan are latecomers and did not start to open their economies until the mid to late 1980s. While financial liberalization can promote the efficiency of resource allocation, it also makes the domestic economy more vulnerable to external economic conditions. Increasing openness of the capital account has reduced the ability of the domestic monetary authorities to conduct monetary policies independent of external considerations. Many countries in the Asian Pacific region - such as Taiwan, Malaysia, and Thailand - have to deal with the monetary consequences of massive capital inflows, even with more flexible exchange rates. They are torn between controlling domestic inflation, on the one hand, and maintaining a competitive real exchange rate, on the other (Glick and Moreno 1994). Because of these problems associated with financial liberalization, most developing countries have been slow to remove various restrictions on capital flows for fear of undermining macroeconomic stability and weakening autonomy in the conduct of monetary and exchange rate policies. These capital controls include strict limits on the inflows and outflows of portfolio investment, narrow foreign exchange exposure ceilings imposed on banks, various kinds of trade restrictions, and administrative controls over interest and exchange rates. Some central banks believe that these measures are necessary and more efficient than sterilization to insulate the domestic economy from external shocks, to retain monetary autonomy, and to stabilize the real exchange rate.

However, the effectiveness of capital controls in developing countries has been widely questioned. By restricting capital flows, the central bank has relegated these flows to the foreign exchange black markets, which have increased the effective degree of capital mobility. Black market erosion of these controls not only has reduced the effectiveness of stabilization policy (Quirk et al., 1987) but also increased the risk of capital flight and the cost of defending the official exchange rate (Gulati 1988). It also has led to currency substitution, which results in loss of seigorage for the government (Agenor 1990). Finally, the effectiveness of restrictive trade policies also is reduced since capital controls usually are needed to maintain such policies. The connection between capital mobility and foreign exchange black markets has been well established in the literature. In the monetary models (Blejer 1978, Agenor 1991) and portfolio-balance models (Dornbusch et al. 1983) of black market exchange rates, foreign exchange is viewed as a financial asset. The demand for foreign exchange can arise from loss of confidence in the domestic currency, inflation tax from holding domestic currency, and low real domestic interest rates. These models have received empirical support (Agenor 1991, Phylaktis and Kassimatis 1994), especially for middle income countries.

Despite the implication of foreign exchange black markets for capital mobility in developing countries, few empirical studies have examined whether these markets have rendered capital controls ineffective. (One exception is Phylaktis (1992) who finds a significant relationship between foreign exchange controls and black market premium for the dollar in Chile.) The answer to this question has important policy implications. First, the effective degree of capital mobility has an important bearing on short-run effects of stabilization policy. Second, if capital controls are ineffective because of black markets, then removing the controls may not exacerbate adverse effects of capital flows on the aggregate economy. Also, capital controls that worked in the past no longer may be effective in dealing with unwanted capital movements and perhaps should be abandoned. Instead, policymakers should devote more attention and resources to "guide" the utilization of capital inflows.

This study attempts to add to the understanding of currency black markets and of capital mobility in developing countries by answering questions as to whether black markets for foreign exchange have significantly weakened capital controls in a sample of Asian Pacific countries - Taiwan, Korea, Thailand, Indonesia, Pakistan, Malaysia, and the Philippines. Currency black markets developed in these countries primarily due to foreign trade restrictions and capital controls. Imposing various forms of trade restrictions such as licensing requirements, quotas, and administrative allocations of foreign exchange creates incentives to smuggle and to fake invoices to satisfy the excess demand for imports. This illegal demand diverts foreign exchange resources in various ways such as underinvoicing exports, over-invoicing imports, other illegal remittances, and smuggling. Coupled with trade restrictions and capital controls, social and political factors also can foster development of black markets - for example, political instability and consequent capital flight under the Marco regime in the Philippines; drug related activities in Thailand; an influx of workers' remittances in Pakistan and the Philippines (Baruni 1989); government corruption, estimated to account for 30% of GDP, in Indonesia (Phylaktis and Kassimatis 1994).


Table 1 presents five measures used in the literature to quantify the degree of capital mobility - (i) covered interest rate parity (CIP), (ii) uncovered interest parity (UIP), (iii) real interest rate parity (RIP), (iv) saving-retention coefficient (Feldstein and Horioka 1980), and (v) the offset coefficient (Argy and Kouri 1974 and Kouri and Porter 1974). Following Frankel (1991) the first four definitions are related and are in descending degree of specificity, with each measure requiring the previous measure to hold plus an additional requirement. The first three measures assume that [TABULAR DATA FOR TABLE 1 OMITTED] international capital flows lead to the convergence of rates of return on financial assets denominated in different currencies. Under CIP, arbitrage activities would equalize the rate of return ([i.sub.t]) of an asset dominated in domestic currency and the rate of return [Mathematical Expression Omitted] of a comparable asset denominated in foreign currency hedged for the foreign exchange risk. The CIP differential [Mathematical Expression Omitted], also known as the country risk premium, captures all barriers to integration of financial markets across national boundaries - such as transaction cost, exchange controls, differential tax treatment, localized information, default risk, and risk of future capital controls.

A somewhat broader measure of capital mobility is UIP under which the rate of return ([i.sub.t]) of an asset denominated in domestic currency is equal to the expected return [Mathematical Expression Omitted] of a comparable asset denominated in foreign currency without forward cover. Expressing the UIP differential, [Mathematical Expression Omitted], as the sum of the CIP differential and an exchange risk premium reveals the relationship between UIP an CIP:

(1) [Mathematical Expression Omitted],

While the country risk premium arises from the difference in the political jurisdictions that issue the domestic and foreign currency assets, exchange risk premium arises from the difference in the assets' currency denomination. The exchange risk premium, measured as the excess of the forward rate over the expected exchange rate, is the extra return that investors demand for assuming foreign exchange risk, which depends on the degree of uncertainty of the exchange rate in the future as well as investors' risk attitude. For instance, if investors are risk neutral the exchange risk premium is equal to zero and for UIP to hold, CIP must hold.

A still broader measure of capital mobility is RIP under which capital flows eliminate real interest rate differentials across countries. The RIP differential can be expressed as the sum of the UIP differential and expected real depreciation:

(2) [Mathematical Expression Omitted]

For RIP to hold, UIP as well as relative purchasing power parity must hold. Therefore, RIP captures both capital market and goods market integration. For instance, RIP differential could exist even without barriers to capital flows as long as purchasing power parity does not hold.

The saving-retention coefficient is the regression coefficient of the national savings ratio regressed against the investment ratios ([[Alpha].sub.2] in table 1). The degree of capital mobility is thought to be inversely related to this coefficient. For example, a low saving-retention coefficient suggests that a decline in national savings could be financed largely by foreign capital inflows, with little increase in the domestic interest rate and consequently no crowding out of investment. According to this logic, perfect capital mobility implies that the saving retention coefficient is zero. Finally, the offset coefficient approach measures capital mobility according to the scope for independent monetary policy when exchange rates are pegged. The scope for independent monetary policy is measured by the fraction, i.e. the offset coefficient, ([[Gamma].sub.2] in table 1) of an increase in domestic credit that is offset by capital outflows. The estimated equation generally takes reduced form from a monetary model as specified in table 1. In this equation, changes in domestic credit (policy variable) by the central bank, nominal income, and interest rate cause changes in net foreign assets (Schadler et al 1993). If capital is mobile, a decrease in net domestic assets will lead to an offsetting increase in net foreign assets. With perfect capital mobility, the offset coefficient is -1.

Empirical research on capital mobility in developing countries has been meager. Most early empirical studies have relied on the offset coefficient approach. These studies typically find that attempts by monetary authorities in developing countries to control the money supply were thwarted by capital outflows, suggesting a high degree of capital mobility (Takagi 1986, Phylaktis 1988). The other four approaches have not been used very often until recently. For instance, Glick and Hutchison (1990) test RIP for a sample of Asian Pacific countries and find that the degree of linkage between the domestic real interest rates and the U.S. real interest rate has increased as financial liberalization in the region has proceeded. On the other hand, Kim (1993) uses the saving-retention approach and finds the saving-retention coefficient to be smaller in countries with more stringent capital controls, suggesting that capital controls have been ineffective in these countries. The lack of empirical research on capital mobility in developing countries is partly due to the lack of relevant data. The nonexistence of forward foreign exchange markets in most developing countries has ruled out the test of CIP. Furthermore, domestic interest rates and exchange rates in developing countries are often subject to administrative controls. This has created an interpretation problem of the empirical results from the testing of CIP, UIP, and RIP. Since black market exchange rates are used in this study, this kind of interpretation problem is less serious. Unfortunately, free market interest rate data from the informal credit markets are available only for Taiwan.

Given the aim of this study, the preferred concept of capital mobility is a narrow one that emphasizes the integration of the financial market. The CIP is the most appropriate choice for this purpose. However, the lack of forward rate data precludes this choice so the next most suitable measure UIP is chosen. Real interest rate parity is not suitable because an investor most likely will not compare the nominal returns from different countries in terms of the expected purchasing power over that country's goods. Instead, all assets will be evaluated by the purchasing power of the goods the investor purchases. The saving-retention measure is not a good choice either as it is a broader measure than RIP and seems to be more relevant to issues relating to balance of payments. Finally, the offset coefficient approach is not suitable because the countries in the sample have adopted a more flexible exchange rate system in recent years. Moreover, this approach involves applying assumptions of a monetary model to balance of payments and hence is not as direct a test for capital mobility as UIP.

Direct empirical studies of UIP for developing countries are scant (see Frankel and Chinn 1993). Edwards and Kahn (1985) propose a variant of UIP for developing countries. They model the domestic market-clearing interest rate (i) as a weighted average of the uncovered interest parity interest rate ([i.sup.UIP]) and the domestic market-clearing interest rate that would be observed if the private capital account were completely closed ([i.sup.C]):

(3) i = [Psi] [i.sup.UIP]+ (1 - [Psi])[i.sup.C] 0 [less than or equal to] [Psi] [less than or equal to] 1

where [i.sup.UIP] is measured as the foreign interest rate adjusted for expected depreciation and [i.sup.C] is measured as the interest rate that would clear the money market when private capital flows are subtracted from the money supply. The parameter [Psi] can be interpreted as an index of the degree of capital mobility. For a closed economy [Psi] = 0 and external factors play no role in the determination of the domestic interest rate. On the other hand, perfect capital mobility would imply a value of one for [Psi]. Although this model has the advantage of setting the degree of capital mobility as an estimable parameter, testing of equation (1) is infeasible because domestic market-clearing interest rates are not available in most developing countries. Haque and Montiel (1991) solve this problem by pointing out that the unobserved domestic market-clearing interest rate should clear the money market. Their finding of a high degree of capital mobility in most developing countries in their sample indicates that the effective degree of capital mobility in these countries is higher than that suggested by the extent of capital controls authorities have imposed. By contrast, Fischer and Reisen (1993) find a low degree of capital mobility for Korea and Taiwan using a similar technique. In any case, one must interpret these results with-caution since they do not constitute a direct test of the arbitrage relationship. Furthermore, the results of this approach could change depending upon how one specifies the counterfactual closed-economy interest rate ([i.sup.C]).


A major problem with testing the uncovered interest parity condition is that exchange rate expectations are unobservable. Cumby and Obstfeld (1981) circumvent this problem by assuming a weak form of market efficiency for the foreign exchange market and examine the time series properties of the exchange rate forecast errors. In a weakly efficient market, investors are "good" forecasters in the sense that they do not make systematic forecast errors based on information from past exchange rate movements. In other words, the exchange rate forecast errors must be equal to zero on the average and are uncorrelated with one another. Mathematically,

(4) [[Epsilon].sub.t+1] = [s.sub.t+1] - [E.sub.t]([s.sub.t+1])

E([[Epsilon].sub.t]) = 0 for all t

E([[Epsilon].sub.t][[Epsilon].sub.t+k]) = 0 for all lags k = 1,2,3,...

where s is the logarithm of the spot nominal exchange rate, E is the conditional expectations operator, and [Epsilon], is the exchange rate forecast error. Hence, market efficiency implies that exchange rate expectation [E.sub.t]([s.sub.t+1]) differs from the realized exchange rate only by a random error. If the UIP condition is true (i.e. [i.sub.t] = [[i.sup.*].sub.t] + [E.sub.t][s.sub.t+1] - [s.sub.t]), then exchange rate forecast error [Epsilon] is

(5) [[Epsilon].sub.t+1] = [s.sub.t+1] + [i.sub.t] - [[i.sup.*].sub.t] - [s.sub.t]

Hence, under the assumption of weak market efficiency, UIP testing involves checking whether the time series [Mathematical Expression Omitted] has a zero mean and is serially uncorrelated. A nonzero mean (that exceeds transaction costs) or the existence of significant autocorrelation would suggest barriers to capital flows. In this paper, we follow this approach to study implications of foreign exchange black markets on capital mobility in developing countries. To this end, the study compares mean and autocorrelations of exchange rate forecast errors from black market exchange rate cases with those from official market exchange rate cases:

(6) [Mathematical Expression Omitted]

(7) [Mathematical Expression Omitted]

where [s.sup.o] and [s.sup.b] are official and black market exchange rates, respectively.

As previously mentioned, a nonzero country risk premium (CRP) or an exchange risk premium (ERP) can account for a UIP differential. More generally,

(8) [[Epsilon].sub.t+1] = [s.sub.t+1]+[i.sub.t] - [[i.sup.*].sub.t] - [s.sub.t] + CRP + ERP

Since the term [s.sub.t+1] + [i.sub.t] - [[i.sup.*].sub.t] - [s.sub.t] is no longer equal to the exchange rate forecast error [Epsilon], in this more general framework, the term UIP forecast error used from now on distinguishes it from the exchange rate forecast error:

(9) UIP forecast error = [[Epsilon].sub.t+1] - (CRP + ERP)

Note that the time series properties of [Epsilon] and the CRP, but NOT those from the ERP can infer the capital mobility concept studied here. A nonzero mean CRP would indicate barriers to capital flows. Similarly, autocorrelations in [Epsilon] would suggest the existence of restrictions to capital flows that prevent investors from exploiting the arbitrage opportunities arising from systematic forecast errors committed in the past. However, all terms of expression (9) are unobservable, except the UIP forecast error. The presence of ERP could complicate the interpretation of the empirical results. For example, a time-varying exchange risk premium that exhibits autocorrelation could induce autocorrelation in the UIP forecast errors. Hence, one must interpret results of this study with care.

Another problem that arises when comparing autocorrelations from two forecast error series is that there is only a point estimate of autocorrelation at each lag. Creating two new random variables [Mathematical Expression Omitted] and [Mathematical Expression Omitted] can circumvent this problem:

(10) [Mathematical Expression Omitted]

(11) [[X.sup.o].sub.t] = ([[UIP.sup.o].sub.t] - [[Mu].sub.o]) ([[UIP.sup.o].sub.t+k] - [[Mu].sub.o])/[[[Sigma].sup.2].sub.o]

where [UIP.sup.b] is the black market UIP forecast error with population mean and variance equal to [[Mu].sub.b] and [Mathematical Expression Omitted] respectively. [UIP.sub.o], [[Mu].sub.o], and [Mathematical Expression Omitted] are the corresponding variables for the official exchange rate case. It becomes obvious that the expected values of [X.sup.b]t and [X.sup.o]t are equal to the k-lag population autocorrelations of the black market UIP forecast errors ([[Rho].sup.b]k) and that of the official UIP forecast errors [Mathematical Expression Omitted] respectively. Therefore, the null hypothesis that [[[Rho].sup.b].sub.k] = [[[Rho].sup.o].sub.k] can be tested by comparing the sample means of the [[X.sup.b].sub.t] and [[X.sup.].sub.t] time series. The t-ratio and degree of freedom for the test (see table 4 for details) are defined under the assumption that the two population variances are unequal (Snedecor and Cochran 1989), as suggested by the results in table 3. We also assume that the two samples are drawn independently.

Sample Periods

Country Sample Period

Malaysia January 1974 - May 1994
Indonesia May 1986 - February 1993
Taiwan November 1980 - May 1994
Pakistan January 1982 - May 1994
Korea January 1980 - May 1994
Thailand November 1984 - May 1994
Philippines 1976:I - 1994:I

Under this assumption, the covariance between [Mathematical Expression Omitted] and [Mathematical Expression Omitted] is equal to zero and [Mathematical Expression Omitted]. This assumption is not only convenient but also makes our statistical inference more conservative. In reality, the two samples are drawn dependently with a positive covariance: a shock that increases the black UIP forecast error is also likely to increase the official UIP forecast error. The sample covariance between [[X.sup.b].sub.t] and [[X.sup.o].sub.t] (not reported) is positive and statistically significant at all lags. Under the independence assumption, the sample variance of [Mathematical Expression Omitted] is overstated, the t ratio is understated, and the chance of not rejecting the null hypothesis increased.


A. Data

The data for this study came from several sources. All black market exchange rate data are from the World Currency Yearbook published by International Currency Analysis of New York. Except for Taiwan, the official market exchange rates and market interest rates are from the International Financial Statistics (IFS line ae and 60b respectively). In the case of Taiwan, the official exchange rate and overnight interbank call rate are from the Financial Statistics Monthly of the Central Bank of China. The empirical analysis uses non-overlapping monthly data in all countries except for the Philippines where it uses quarterly data. Table 2 presents the sample periods. Selecting sample periods excluded fixed exchange rate periods to avoid the so-called "peso" problem (see Lizondo 1983) often associated with studying currencies under a fixed exchange rate system. For example, under a fixed exchange rate system, the domestic interest rate will reflect a small probability of a devaluation. As long as the devaluation does not occur, [i.sub.t] - [[i.sup.*].sub.t] + [s.sub.t] will overestimate the future exchange rate and the exchange rate forecast errors will exhibit positive autocorrelation even if the market is efficient.

B. Results and Discussion

Table 3 presents summary statistics of the black market and official market UIP forecast errors. Column [t.sub.[Mu]] reports the test results of the null hypothesis that their means are equal while the column [F.sub.[Sigma]] tests the null hypothesis that their variances are equal. Several points are worth noting. First, except for Malaysia and Pakistan, the mean black market and official market UIP forecast errors are statistically significant. Second, the mean UIP forecast errors under the two exchange rates are very close, as the [t.sub.[Mu]] column shows. The reason may be as follows. On the one hand, since foreign exchange black markets provide a way to circumvent capital controls, which are probably the most important impediment to financial market integration, the country risk premium in the black market cases should be less than that in the corresponding official market cases. On the other hand, to the extent that black market exchange rates are more volatile and black market activities carry the risk of punishment, the exchange risk premium in the [TABULAR DATA FOR TABLE 3 OMITTED] black market cases should be greater than that in the corresponding official market cases. Third, consistent with the market-determined nature of black market exchange rates, variances of the black market UIP forecast errors are greater than those of the official market UIP forecast errors. The [F.sub.[Sigma]] reported in the last column shows that the differences are statistically significant for all countries except Malaysia and Indonesia. Since these two countries have the most open capital account in the sample, it is not surprising that their official and black market exchange rates exhibit similar degrees of volatility.

Table 4 reports autocorrelations for the first 10 lags of the two UIP forecast errors, and figures 1 to 7 plot the corresponding autocorrelation functions. Most of the autocorrelations are statistically significant in the first three lags. The Ljung-Box statistics (not reported) are significant at all lags for all countries. The figures show that the autocorrelations of the official UIP forecast errors are greater than those of the black market UIP forecast errors at all lags. The last column [t.sub.[Rho]] of table 4 reports the t test of the null hypothesis that the k-lag autocorrelation of the black market UIP forecast errors differs significantly from that of the official UIP forecast errors. The results show that the gaps between the two autocorrelation functions are statistically significant at most lags for the four countries with more stringent capital controls, which in descending order of degree of capital controls, are Korea, Taiwan, Thailand, and Pakistan. Note that the degree of statistical significance is directly proportional to the degree of capital control, with the [t.sub.[Rho]]s for Taiwan and Korea more significant than those of Pakistan and Thailand. On the other hand, the gap between the official and black market autocorrelation functions is not significant for Malaysia, Indonesia, and the Philippines, where capital controls are much less restrictive.


If the two risk premiums do not exist, the above results provide evidence that in the countries with more stringent capital controls, the currency black markets have significantly reduced the effectiveness of these controls and increased the effective degree of capital mobility. The greater autocorrelations under the official exchange rate cases would reflect a more autocorrelated exchange rate forecast error [Epsilon]. Hence, capital controls in the official market result in systematic forecast errors committed by investors.

Presence of the two unobservable premiums does not seem to affect the conclusions. For the autocorrelation of the country risk premium, it is unlikely that there is a significant difference between the official and black exchange rate cases. The reason is that the country risk premium captures mainly barriers to capital flows due to institutional and policy-induced factors - such as transaction costs, tax regulation, information costs, and enforcement of capital control regulation. These factors tend to be relatively stable and change slowly with time. As for the exchange risk premium, it is difficult to know whether the autocorrelation in the black market cases is greater or less than that in the official market cases. However, the magnitude of the premium should be much greater in the black market cases due to the greater degree of black market exchange rate volatility. Therefore, the contribution of the black exchange risk premium to the autocorrelation of the UIP forecast error should be greater than that from the official exchange risk premium. This conjecture makes the results reported in table 4 as well as figures 1 to 7 all the more remarkable. They suggest that black market exchange rates did respond to yield divergence between domestic and foreign assets and increase the degree of capital mobility in countries with stringent capital controls. The significant downward shift of the autocorrelation functions of these countries points to the futility of capital controls. This confirms Baruni's (1989, p. 220) interesting observation that "if the government wishes to retain its power, it has to be willing not to use it to the fullest extent."


Many developing countries traditionally have used capital controls to preserve foreign exchange reserves, support an overvalued currency, control the money supply and domestic inflation, and channel resources to "priority sectors" in order to foster economic growth. They view these controls as an effective policy instrument to deal with the undesirable effects of financial opening. This paper examines whether foreign exchange black markets have significantly eroded the effectiveness of capital controls and contributed to an increase in capital mobility in a sample of Asian countries. The analysis tests the uncovered interest parity condition using the official market and black market exchange rates. Despite the fact that UIP forecast errors under the black market rates are more volatile than those under the official exchange rates, their autocorrelations are significantly smaller than those under the official exchange rates in countries known to have more stringent capital controls. Therefore, the movements of black market exchange rates in these countries seem to have responded to arbitrating opportunities arising from yield differentials between domestic and foreign assets, suggesting that these clandestine markets have reached a level of size and maturity to threaten the effectiveness of existing capital controls. This conclusion echoes results by Kim (1993) that the extent of capital controls has relatively little bearing on the saving retention coefficient. If exchange controls are ineffective anyway, the logical conclusion is to speed up the pace of dismantling them and let prices reflect the scarcity of foreign exchange. The existence of currency black markets is socially wasteful because participants in these markets spend a large amount of resources to avoid punishment and in rent seeking behaviour, while the government wastes resources in futile attempts to police a system that is not enforceable and loses its credibility in the process. Furthermore, recent research has suggested that the real exchange rate price distortions created by black markets have exerted a negative impact on economic growth (Gyimah-Brempong 1991, Gyimah-Brempong and Gyapong 1993). By relaxing capital controls, the resources saved from enforcing ineffective capital controls could be released to develop other ways to deal with capital flows. These include enforcing sound macroeconomic policies that would eventually attract foreign capital to long term investment projects and developing market instruments to sterilize capital inflows. The Employee Provident Fund in Malaysia and the Bank Indonesia Certificates in Indonesia are examples of market institutions that could manage problems from capital flows more effectively than can direct capital controls.


CIP: covered interest parity RIP: real interest rate parity UIP: uncovered interest parity

This is a revised version of a paper presented at the Western Economic Association International 70th Annual Conference, San Diego, CA, July 7, 1995, in a session organized by Professor Jack W. Hou, California State University, Long Beach and Dr. Baekin Cha, City University of Hong Kong. The author thanks Dr. Yu Ming Fu, Dr. C. Y. Sin, Professor Lilian Ng, and two anonymous referees for helpful comments and suggestions. The author gratefully acknowledges financial assistance provided by the City University of Hong Kong through grant 903-372.


Agenor, Pierre-Richard, "Stabilization Policies in Developing Countries with a Parallel Market for Foreign Exchange: A Formal Framework," International Monetary Fund Staff Papers, 37, September 1990, 560-592.

-----, "A Monetary Model of the Parallel Market for Foreign Exchange," Journal of Economic Studies, December 1991, 4-18.

Argy, Victor and Pentti Kouri, "Sterilization and the Volatility in International Reserves," in National Monetary Policies and the International Financial System, ed. by Robert Z. Aliber, Chicago: University of Chicago Press, 1974.

Baruni, Tariq, "Black Markets, Openness, and Central Bank Autonomy," World Institute for Development Economic Research, Working Paper No. 62, August 1989, Helsinki, Finland.

Beljer, Mario I., "Exchange Restrictions and the Monetary Approach to the Exchange Rate", in The Economics of Exchange Rates, ed. by Jacob A. Frenkel and Harry G. Johnson, Reading: Addison-Wesley, 1978, 117-128.

Cumby, R. and M. Obstfeld, "A Note on Exchange-Rate Expectations and Nominal Interest Differentials: A Test of the Fisher Hypothesis," Journal of Finance, June 1981, 697-703.

Dornbusch, R., Daniel Dantas, Clarice Pechman, Roberto Rocha, and Demetri Simoes, "The Black Market for Dollar in Brazil," Quarterly Journal of Economics, February 1983, 25-40.

Edwards, S. and M. S. Khan, "Interest Rate Determination in Developing Countries: A Conceptual Framework," IMF Staff Papers, September 1985, 377-403.

Feldstein, Martin and C. Horioka, "Domestic Saving and International Capital Flows," Economic Journal, June 1980, 314-329.

Fischer, B. and Helmut Reisen, Liberalizing Capital Flows in Developing Countries, OECD, 1983.

Frankel, Jeffrey, "Quantifying International Capital Mobility in the 1980s'," in Berntheim, D. and Shoven, J. eds., National Saving and Economic Performance, Chicago: University of Chicago Press, 1991, 227-269.

Frankel, Jeffrey and M. Chinn, "Financial Links Around the Pacific Rim: 1982-1992," Pacific Basin Working Paper Series, 93-08, Federal Reserve Bank of San Francisco, April 1993.

Glick, Reuven and M. Hutchison, "Financial Liberalization in the Pacific Basin: Implications for Real Interest Rate Linkages," Journal of Japanese and International Economies, 4, 1990, 36-48.

Glick, Reuven and Ramon Moreno, "Capital Flows and Monetary Policy in East Asia," Pacific Basin Working Paper Series, Working Paper 94-08, Federal Reserve Bank of San Francisco.

Gyimah-Brempong K., "Effects of Exchange Controls on Black Market Exchange Rate in a Less Developed Country," International Economic Journal, Spring 1991, 63-78.

Gyimah-Brempong K. and A. Gyapong, "Exchange Rate Distortion and Economic Growth in Ghana," International Economic Journal, Winter 1993, 59-73.

Gulati, Sunil, "Capital Flight: Causes, Consequences, and Cures," Journal of International Affairs, Fall 1988, 165-185.

Haque Nadeem U. and Peter Montiel, "Capital Mobility in Developing Countries: some Empirical Tests," World Development, 19:10, 1991, 1391-1398.

Kim, Sun Bae, "Do Capital Controls Affect the Response of Investment to Saving?" Economic Review, Federal Reserve Bank of San Francisco, 1, 1993, 23-39.

Kouri, Pentti and M. Porter, "International Capital Flows and Portfolio Equilibrium," Journal of Political Economy, May/June 1974, 443-467.

Larue, Bruno and J. Mutunga, "The Gavity Equation, Market Size, and Black Market Exchange Rates," International Economic Journal, Summer 1993, 6175.

Lizondo, J. S., "Interest Differential and Covered Arbitrage," in Financial Policies and the World Capital Market: The Problem of Latin American Countries, ed. by Pedro Aspe Armella, Rudiger Dornbusch, and Maurice Obstfeld, Chicago: University of Chicago Press, 1983, 221-240.

Phylaktis, Kate, "Capital Controls: The Case of Argentina," Journal of International Money and Finance, September 1988, 303-320.

-----, "Black Market for Dollar in Chile," Journal of Development Economics, November 1991, 155-172.

Phylaktis, K. and Y. Kassimatis, "Black and Official Exchange Rates in the Pacific Basin Countries: An Analysis of Their Long-run Dynamics," Applied Economics, April 1994, 399-407.

Quirk, Peter J., and others, "Floating Exchange Rates in Developing Countries: Experiences with Auction and Interbank Markets," International Monetary Fund Occasional Paper, 53, Washington, D.C., 1987.

Schadler, S., M. Carkovic, A. Bennett, and R. Kahn, "Recent Experience with Surges in Capital Inflows," International Monetary Fund Occasional Paper, 108, Washington, D.C., 1993.

Snedecor, George W. and W. Cochran, Statistical Methods (8th ed), Ames, Iowa: The Iowa State University Press, 1989.

Takagi, Shinji, "Rediscount Policy and Official Capital Flows: A Study of Monetary Control in Central America in the 1950s," Canadian Journal of Economics and Political Science, November 1986, 475485.

Wong: Assistant Professor, Department of Economics and Finance, City University of Hong Kong, 83 Tat Chee Road, Kowloon Tong, Hong Kong 852-2788-7948, Fax 852-2788-8806
COPYRIGHT 1997 Western Economic Association International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1997 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Wong, Clement Yuk Pang
Publication:Contemporary Economic Policy
Date:Jan 1, 1997
Previous Article:Binge drinking in college: the impact of price, availability, and alcohol control policies.
Next Article:Optimum currency areas and European monetary unification.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |