An Empirical Analysis on Capital Flows: The Case of Korea and Mexico.

• Author: Ying, Yung-Hsiang

Yung-Hsiang Ying [*]

Yoonbai Kim [+]

This paper discusses causes of capital flows in Korea and Mexico. Both countries received substantial amounts of foreign capital in the late 1980s and early 1990s. International capital helped these countries achieve a higher standard of living and faster economic growth. However, undesirable macroeconomic effects such as appreciation of real exchange rate and widening current account deficits usually accompany foreign capital inflows. The vector autoregressive (VAR) method is applied to investigate the underlying shocks causing the capital inflows. The main findings are that the U.S. business cycle and shocks to foreign interest rates account for more than 50% of capital inflows to both countries in the past two decades.

1. Introduction

   After a slow period during the 1980s, the flow of international capital into developing countries sharply increased during the 1990s. Between 1990 and 1994, countries in Asia and Latin America received capital inflows of $670 billion (Calvo, Leiderman, and Reinhart 1996). The experience of developing countries suggests that increased movement of international capital is a mixed blessing. On the one hand, funds borrowed from abroad facilitate investment and stimulate economic growth in recipient countries. On the other hand, capital inflows are often accompanied by an appreciation of the real exchange rate, a widening of the current account deficit, and an increase in inflationary pressures. The appreciation of the real exchange rate decreases the competitiveness of the trade sector and increases the vulnerability of the banking system in capital-recipient countries. Furthermore, the reversal of capital movements can be very painful, as vividly shown in cases of the Mexican foreign exchange crisis in late 199 4 and the financial crises in Southeast and East Asian countries at the end of the 1990s.

   International movements of capital can be motivated by factors within recipient countries and by factors external to the recipient countries. Internal factors include fiscal reform and deficit reduction, inflation stabilization, successful resolution of debt crisis, trade liberalization, and dismantling of restrictions on capital flows. Among external factors, low interest rates and sluggish growth in the United States and other industrial countries tend to encourage investors to shift resources to the emerging markets. During the late 1980s and early 1990s, changes in many of the above factors provided favorable conditions for foreign capital in developing countries. Calvo, Leiderman, and Reinhart (1993) argue that low interest rates in the United States motivated massive capital movements to Latin America. Bohn and Tesar (1996) studied the differences in the timing of investment in individual countries and concluded that domestic factors were more important than external factors in Asian countries. [1]

   Empirical evidence suggests that the effects of international capital movements tend to differ depending on the types of capital flows. Studies by Dooley (1988), Goldstein, Mathieson, and Lane (1991), and Edwards (1991) find that capital flows are accompanied by fewer problems when they are long-term direct investment induced by the domestic growth prospects of the recipient countries. However, capital flows that take the form of portfolio investment and originate from external shocks (such as changes in foreign interest rates) tend to be short term and increase the difficulty of economic stabilization in capital-recipient countries. [2]

   Empirical studies before the financial crisis in Asia have contrasted capital flows to Latin America and those to Asia. They generally argue that capital flows to Asia were of a higher quality than the capital flows to Latin America, as the former consisted primarily of direct investment while the latter consisted primarily of portfolio investment. The studies also note that macroeconomic policies have been more stable in Asia than in Latin America. Now that we know Asian countries can experience financial crises, it would be interesting to review the causes of capital flows to Asia and Latin America.

   The purpose of this study is to investigate the macroeconomic factors of capital flows in Korea and Mexico. Both countries received large volumes of foreign capital. In both countries, the period characterized by capital inflows and booming domestic economies came to an abrupt end with the beginning of financial crisis.

   In this study, we apply a structural vector autoregressive (VAR) method that relies on long-run restrictions to identify structural shocks to the economy. We consider both internal and external causes. Determining the relative roles of internal and external factors in driving capital flows is not only theoretically important but also a crucial issue for policymakers. To the extent that internal factors are important, domestic policymakers may have more leverage on capital flows by sound macroeconomic policy. If capital flows are dominated by external factors, however, capital flows may be more difficult to control.

   The content of this paper is as follows. The next section presents the data on capital flows. An empirical model of capital flows in a small economy is developed in section 3. Estimation results are presented in section 4. The paper concludes with discussion and policy implications in section 5.

2. Capital Flows in Korea and Mexico

   In the presence of central bank intervention in the foreign exchange market, the sum of the current and capital account balances is not zero. Instead, the sum, which is also called the balance of payments, leads to changes in foreign reserve holdings at the central bank. When it is positive or when there is a surplus in the balance of payments, the central bank purchases foreign assets in net. Data on international transactions are difficult to find and typically short in length. To obtain longer series of data, we approximate the current account balance by the trade balance (net exports of merchandise) for Mexico or by net exports of goods and services for Korea. The use of the proxies should not distort statistical inference in any qualitative manner. Two justifications can be offered for our decision: (i) The current account balance is dominated by merchandise trade balance in the two countries, as in most other countries; and (ii) the correlation between our proxy and the correctly measured current accou nt balance, when both are available, is very high: 0.988 for Korea and 0.945 for Mexico. Figure 1a and b compares the behavior of the current account balance (dash lines) and the trade balance (thick lines). They make it clear that during the periods for which we have data for both series, the two move quite closely, as indicated by the high correlation coefficients.

   The overall balance is obtained as the change in foreign reserve holding at the central bank during the period. The capital account balance is then obtained by subtracting the current account balance from the overall balance. Both current account and capital account balances are in domestic currency units. Both are expressed in real terms by dividing by the wholesale price level. Figure 2a and b depicts the balances on the current and capital accounts scaled by gross domestic product from 1960:1 to 1996:4 for Korea and from 1981:1 to 1996:4 for Mexico. [3] Current account and capital account balances are denoted as the thick and dash lines, respectively.

   Evidently, capital flows in the recent period have become larger and more volatile. In Mexico, surges in inflow of the late 1980s and early 1990s and a drastic reduction in capital inflows in the mid-1980s and 1994 are noteworthy. In contrast, Korea experienced a steady increase in capital inflows beginning in the mid-1980s. Figure 2a and b also indicates that the current and capital account balances are near mirror images to each other until the first half of the 1980s in both countries. After that, these two series deviate from each other. There was no similar reduction in capital inflow (or increase in capital outflow) to match a surge in current account surplus in Korea during the late 1980s. Similarly, a sharp reduction in capital inflow in Mexico after 1994 was matched by a decrease in current...