An empirical analysis of outward foreign direct investment of Korea and Japan

Bae, Sung C

This paper examines the empirical relationship between Korean and Japanese outward foreign direct investment (FDI). Considering the heavy dependence of Korean firms on Japanese technology and the intensive competition of Korean firms against Japanese firms, it is argued that Korean firms’ FDI is affected by Japanese movement for outward FDL The results of this paper provide evidence supporting the significant positive relationship between Korean FDI and Japanese FDL Specifically, current year’s Korean FDI is significantly positively related with previous year’s Japanese FDI even after the effects of other macroeconomics variables on the Korean FDI are controlled. This paper also offers interesting findings on the relationships between macroeconomic variables and Korean FDL In particular, greater Korean FDI is found to be associated with an increase in Korean wages relative to Japanese wages and with an appreciation of Japanese yen against Korean won.

INTRODUCTION

This paper focuses on foreign direct investment (hereafter, FDI) activities of the two more advanced economics in Pacific-Basin countries, Japan and Korea. Japan and Korea exhibit some similarities in their FDI activities during the past decade. During the 1980s, Japanese exporting industries were expanding globally to remain competitive with both foreign and domestic companies. During the late 1980s and early 1990s, however, Japanese firms experienced severe economic downturns and restructurings following the so-called “bubble economy.” Combined with these economic conditions domestically, according to Komiya (1991), deregulation of financial sectors in Japan and increased import barriers by major trading partners have contributed to the increase in Japan’s FDI.

During the 1980s, Korean industries faced similar economic hardships. Along with increased pressure from major trade partners, including the U.S. and Western Europe, to open its economy, domestic problems such as wage hikes, labor shortage, frequent labor disputes and appreciation of Korean won= have led Korean firms to invest abroad. The increase in outward FDI by Korean firms could also be attributed to changes in global markets including intensifying protectionism in trade and technology, formation of economic blocs, and incentives offered by many overseas countries (e.g., China, Indonesia and Mexico). For the first time, FDI by Korean Firms exceeded foreign investment in Korea in 1990 [Kim 1993]. While previous literature has identified several important determinants of a country’s outward FDI, this paper investigates the empirical relationship in outward FDI activities between Korea and Japan. In particular, it is argued in this paper that Korean FDI is further influenced by Japanese FDI in addition to the macoeconomic variables documented as determinants of outward FDI by previous studies. Traditionally, Korean and Japanese firms have not only worked as partners but also competed against each other in the world markets. For example, Korean firms have competed with Japanese firms in the U.S. markets mainly for automobiles, electronics, and computer products. Firms in both countries also compete against each other in other lessdeveloped Asian countries including China, mainly to take advantage of cheap labor and local production costs. Considering the heavy dpendence of Korean firms on Japanese technology and the intensive competition of Korean firms against Japanese firms, it is expected that Korean firms’ FDI is strongly positively related to Japanese to Japanese firms’ FDI.

In order to investigate the empirical relationship between Korean and Japanese FDI, regression analysis is performed on FDI data for 1977-92. As a country’s outward FDI is supposedly influenced by macroeconomic variables, such as GDP growth rate, interest rate, trade balance, exchange rate, and wage growth [United Nations 1992], these variables are used as control variables in regressions of Korean FDI, with Japanese FDI as the test variable.

The results of this paper provide strong evidence supporting the close relationship between Korean and Japanese FDI. After controlling for the effects of other macroeconomic variables on FDI, current year’s Korean FDI is found to be significantly positively related with previous year’s Japanese FDI. This paper further provides additional empirical insights into the effects of several macroeconomic variables for Korea relative to Japanese Korean FDI.

RELATED STUDIES AND DEVELOPMENT OF HYPOTHESIS

Several previous studies have examined determinants of a country’s outward FDI, focusing mainly on Korea and Japan [Hennart and Park 1994; Lee 1994; Lucas 1993; Kim 1993; Mann 1993; Serapio 1993; United Nations 1992]. These studies have identified as important determinants of a country’s outward FDI: GDP growth rate, interest rate, trade balance, exchange rate, and wage growth rate. In this paper, it is argued that Korean outward FDI is further influenced by Japanese outward FDI in addition to these macroeconomic variables.

Historically, Korean and Japanese fins have not only worked as partners but also competed against each other. Korean industries have been structured as export-oriented in a similar manner to Japanese industries. Despite Korea’s efforts to attract European and U.S. investment since the late 1960s, the economic growth of Korea has depended largely on Japanese investment.3 Based on the capital and technology from Japan, Korea has developed many export industries including steel, shipbuilding, textiles, and, in recent years, automobiles and electronics [Matsuura 1989]. During the late 1980s, faced with the so-called `three highs’ -high labor costs, high borrowing costs, and high Korean won value-Korean firms realized that they were no longer cost competitive, especially against Japanese products. This was when Korean firms resorted to FDI in other less-developed Asian countries, mainly to take advantage of cheap labor and to remain cost-competitive. Hence, it is expected that the heavy dependence of Korean firms on Japanese technology and the intensive competition of Korean firms against Japanese firms have made Korean firms shape their FDI policies, at least in part, as defensive tactics against Japanese firms. Hence, Korean FDI is expected to be significantly influenced by Japanese FDI in the same direction; a positive relationship between Japanese FDI and Korean FDI is hypothesized.

DATA AND RESEARCH METHODOLOGY

Data

The data on FDI and other economic and financial information were collected from various issues of International Financial Statistics published by International Monetary Fund, Korea and the World; Key Indicators and Foreign Trade Trends and Key Indicators published by Korea Foreign Trade Association and Japan: An International Comparison published by Keizai Koho Center. Additional information on FDI was also obtained from various issues of Business Korea magazine. Standard and Poor’s Global Vantage CD-ROM database was also reviewed to collect data on exchange rates and other economic indicator variables.

Table 1 reports the approval-basis outward FDI amounts of Korea and Japan by three host country regions-North America, Europe, and Asia-during the 1977-92 period. The total amount of Japanese FDI is far greater than that of Korean FDI; in 1992 alone, Korean FDI is only 3.6 percent of Japanese FDI ($1.218 billion vs. $34.138 billion). Over the entire period, however, Korean FDI increased dramatically by 66.7 times (1,218/18 – 1), while Japanese FDI increased by 11.2 times (34,318/2,806 – 1). The sharp increase in Korean FDI can be attributed to changes in locational advantages related to its economic development. As indicated earlier, wages rose rapidly in Korea, and trade restrictions increased strongly from its major partners; between 1985 and 1992, labor productivity in manufacturing increased 92 percent but wages increased by 196 percent in the same period. Further, the Korean won appreciated from 870.02 won to 780.65 won (21.8 percent) against the U.S. dollar during the same period [Korea Foreign Trade Association 1993].

Geographically, Asia is the most important market for Korean FDI followed by North America and Europe; in 1992, 47.9 percent of Korean FDI was made in Asian countries, compared to 18.8 percent of Japanese FDI made in the same region. It is also shown that Japan is far agead of Korea in its investment in the U.S. and Europe with respect to both the dollar amounts and the proportion of FDI; these two regions represent 63.4 percent and 44.0 percent of FDI for Japan and Korea, respectively.

Regression Model

Previous studies have documented that a country’s outward FDI is influenced by several market and economic variables including the country’s GDP growth, interest rates, trade balance, labor costs, and currently rate [see United Nations 1992 for a review].

In order to examine the hypothesized relationship between Korean FDI and Japanese FDI, the following time-series regression model of Korean FDI is run against Japanese FDI is run against Japanese FDI with six macroeconomic variables as control variables.

KFDI^sub t^=aO+(alpha ^sub 1^ L(KGDPG)^sub t^+(alpha)^sub 2^L(KGBD)sub t^ +(alpha)^sub 3L(KTRADE)^sub t^+(alpha)^subL(WDOL)^sub t^ +(alpha)^sub 5^L(WYEN)^sub 6^+(alpha)^sub 6^L(KWAGEN)^sub t^ +(alpha)^sub 7^L(JFDI)^sub t^+e^sub t^ (1) Where L( ) denotes the lag operator such that, for example, L(KWAGE) is KWAGE lagged one year. As a country’s FDI activity in a given year is a response to the outcome of export activities and economic conditions in the previous year, an analysis of regressing current year’s Korean FDI against previous year’s Japanese FDI and other macroeconomic variables would provide more meaningful evidence. Hence, the regression model of Korean FDI is estimated with lagged independent variables.4 In Equation (1), all variables except for JFDI are measured for Korea during the period of 1977-1992.

Description of Variables

Dependent Variable

KFDI represents the annual total amounts of outward foreign direct investment of Korean firms during the 1977-92 period. Both the raw values of KFDI and two adjusted versions of KFDI are used in the regression analysis; KFDI divided by Korea’s annual GDP and KFDI divided by Korea’s annual total investment.5

Control Variables

KGDP represents the annual growth rate of gross domestic product (GDP) in Korea. KFDI is expected to be positively related to KGDPG. As the GDP growth rate in the previous year increases, a market attains more size that permits local production. Accordingly, the market becomes more saturated by increased local competition. Hence, local firms will be more inclined to invest in foreign countries to avoid increased local competition. This notion is supported by previous studies of U.S. FDI in the European Community [see, e.g., Bandera and White 1968; Scaperlanda and Mauer 1973; Schmitz and Bieri 1970; Lunn 1980; and Scaperlanda and Balough 1983]. Torrisi (1985) documents similar evidence on U.S. FDI in Columbia.

KGBD represents yields on long-term government bonds in Korea. KFDI is expected to be positively influenced by KGBD. As an interest rate increases, a firm’s cost of capital increases. Since local fims become less competitive due to the increase in cost, they are expected to be more outward oriented. Petrochilas (1984) analyzes time-series data for FDI in Greece and finds that a country’s discount rate, lagged one year, is negatively, and often significantly, related to its FDI. These findings demonstrate the relevance of a source of complementary local capital as a location advantage for FDI. As another measure of the interest rate, KPRIME is also used, which represents the prime interest rate in Korea.

KTRADE represents the trade balance in Korea. KFDI is expected to be negatively influenced by KTRADE. A trade surplus may indicate that local companies have a competitive advantage for their product in the world market; hence, local firms would prefer to remain status quo.6 On the other hand, a trade deficit may indicate that local firms are less competitive due to the difficulty of export; hence, the local firms are expected to be more outward oriented.

WYEN and WDOL represent the exchange rate of Korean won against Japanese yen and U.S. dollar, respectively. As the value of Korean won appreciates, so does the price of exported products. Hence, Korean product will be less competitive in the world market, and, accordingly, local companies will be more likely looking outward for new production facilities overseas. This suggests a negative relationship between WYEN (WDOL) and KFDI.

KWAGE represents changes of nominal wages in Korea. Korean FDI is expected to be positively influenced by wage cost. As wage cost increases, the cost of product will increase. Hence, the local firms will be less competitive. In face of competition from foreign products, the firms will be more inclined to invest overseas where labor costs are cheaper. Reidel (1975) analyzes export-oriented FDI in Taiwan Province of China. His study finds a significant positive relationship between the magnitude of FDI and the difference in efficiency-wage between Taiwan Province of China and key investing economies including Hong Kong, Japan, and the U.S.7 Lall’s (1980) study of U.S. FDI offers similar evidence that the average wage per employee in the U.S. tends to be positively related to overseas investment by U.S. firms.

Test Variable

JFDI measures the annual amounts of FDI by Japanese firms during the 1977-92 period. As discussed in the previous section, a positive relationship between JFDI and KFDI is expected due to the heavy dependence of Korean firms on Japanese technology and the intensive competition of Korean firms against Japanese in the world market. To be consistent with three different measure of the dependent variable, both the raw values and two adjusted (by Japanese GDP and Japanese annual total investment) versions of JFDI are employed.

EMPIRICAL RESULTS

Regression Results

Table 2 presents the regression results. Three separate regressions were performed with a different independent variable in each regression. Regression 2 uses the unadjusted, raw values of KFDI, and Regressions 2 and 3 use the adjusted KFDI, divided by Korean GDP and Korean total investment, respectively. All three regressions take lagged values of KGDPG, KGBD, KTRADE, WDOL, WYEN, KWAGE, and JFDI as the independent variables to measure the effect of these lagged variables on KFDI. The Durbin-Watson statistics in all regressions assure no autocorrelation between the successive values of residual terms.

The overall results reported in Table 2 regarding independent control variables generally are consistent with those predicted. For all the regressions, the estimated regression coefficients for L(KTRADE) have the expected negative signs, all significant at the 0.05 level. Thus, a greater trade balance in the previous year is associated with greater KFDI in the current year. L(KWAGE) have also the expected positive signs and all the significant at the minimum 0.05 level. Hence, an increase in wage costs in a given year appears to significantly contribute to an increase in Korean firms’ outward FDI in the following year. The estimated regression coefficients for L(KGDPG) have the expected signs, but are not significant at the 0.10 level. On the contrary, the regression coefficients for L(WYEN) carry significant positive signs (at the 0.05 level) in all regressions. This result provides support for the notion that as a depreciation of Korean won against Japanese yen (or equivalently, an appreciation of Japanese yen) in a given year makes the Japanese technology and product imported by Korean firms more expensive, Korean firms increase their overseas investment in the following year to remain cost-competitive.8

Our primary focus is on the regression coefficients on L(JFDI). As hypothesized, the estimated regression coefficients for L(JFDI) have the expected positive signs and are all significant at the 0.05 level. Hence, it is evident that current year’s Korean FDI is strongly influenced by previous year’s Japanese FDI even after controlling the effects of other macroeconomic determinants on Korean FDI.

Further Analysis of Korean and Japanese FDI

So far the analysis have considered the macroeconomic variables associated with Korean economy as determinants of Korean FDI. While these variables in Korea explain to a great extent the activities of Korean FDI, it is also conceivable that the magnitude of each variable in Korea relative to the same variable in Japan would also affect the magnitude of Korean FDI relative to that of Japanese FDI. In order to examine the relative influences of macroeconomic variables between Korea and Japan, the following regression model is performed with the ratio of current year’s Korean FDI to previous year’s Japanese FDI, KFDI/L(JFDI), as a dependent variable.

KFDI^sub t^/LJFDI^sub t^=Beta ^sub 0^+Beta^sub 1^L(R(GDPG))^sub t^ +Beta^sub 2^L(R(GBD))^sub t^ +Beta ^sub 3^L(R(TRADE))^sub t^ +Beta^sub 4^L(WYEN)^sub t^ +Beta^sub 5^L(R(WAGE))sub t^+epsilon^sub t^ (2) where L(.) denotes a lag operator and R(.) denotes a ratio operator such that, for example, L(R(GDPG)) is measured as the ratio of Korean GDPG to Japanese GDPG in the previous, or L(KGDPG)/L(JGDPG).

Table 3 represents regression results estimated from equation (2). Two regressions were performed, with each regression employing either unadjusted KFDI and L(JFDI) or adjusted KFDI and L(JFDI) by each country’s GDP as the dependent variable.9

Several interesting results are obtained. Among others, the estimated regression coefficients for L(R(WAGE)) have positive signs, both significant at the 0.01 level. The positive coefficient for L(R(WAGE)) indicates that a greater wage increase in Korea relative to that in Japan in the previous year is associated with greater KFDI in the current year. Since the Korean firm products compete directly with Japanese products in many industries, a higher increase in wages in Korea relative to wages in Japan will make Korean products less competitive than Japanese products. Hence, in order to remain competitive with respect to labor cost, Korean firms are expected to increase their outward FDI. Consistent with the findings in the previous section, the estimated coefficients for L(WYEN) are positive and significant at the 0.01 level. This result confirms the significant positive effect of the value of Korean won relative to that of Japanese yen. The appreciation of Japanese yen makes the imported Japanese technology and product by Korean firms more expensive and, consequently, Korean firms are more inclined to invest overseas (mainly in less-developed Asian countries) to secure less-costly resources.

SUMMARY AND CONCLUSIONS

This paper has examined the causal relationship between Korean FDI and Japanese FDI. During the late 1980s and early 1990s, combined with the deteriorating economic conditions domestically, the deregulation of financial sectors in Japan and increased import barriers by major countries have contributed to the increase in Japan’s FDI. During the 1980s, along with increased pressure from major trade partners, domestic problems such as wage hikes, labor shortage, frequent labor disputes and appreciation of the Korean won have led Korean companies to invest abroad.

Considering the heavy dependence of Korean firms on Japanese technology and the intensive competition of Korean firms against Japanese firms in the world market, it is argued that Korean firms’ FDI would be strongly influenced by Japanese movement for outward FDI. The results of this paper provide support for a strong positive relationship between Korean FDI and Japanese FDI. The results show that previous year’s Japanese FDI strongly influences current year’s Korean FDI even after the effects of several macroeconomic variables on Korean FDI are controlled. This paper also offers several interesting results on the relationships between Korean FDI and these macroeconomic determinants. In particular, an increase in FDI by Korean firms is associated with high Korean wages relative to Japanese wages relative to Japanese wages and with a greater appreciation of Japanese yen again Korean won.

Two potential issues deem warranted for further analysis. First, the recent changes in FDI by Korea and Japan have been a sector-oriented restructuring strategy, i.e., the changes were found in a few industries. For example, the surge of FDIs by Korea’s small- and mediumsized firms since the late 1980s was made mainly in light, labor intensive industries toward the newly industrializing Asian countries. Hence, an analysis of FDIs at the industry level would provide additional insights into the outward FDI behavior of firms in these countries. Second, there are two roles of Japanese FDI. Once is the role of providing capital and technology to Korea. The other is the role of Japanese FDI in third countries that competes with Korean exports. It is possible that, because Japanese and Korean industries are so similar, given the considerable transfer of capital and technology from Japan to Korea, they are both affected in similar ways with respect to outward FDIs. Hence, more meaningful evidence could be documented by distinguishing these two roles of Japanese FDI.

NOTES

The paper was completed when Hwang was at the Graduate School of Business, Bowling Green State University. Bae acknowledges resulearch support from the Ohio Board of Regents Research Challenge Grant Program.

1. This resulted mainly from the trade surplus in Korea during the 1986-1989 period.

2. Japan has been major trade partner of Korea during the last five decade. In 1992, trade with Japan represented 15.1 percent of total exports (ranked second) and 23.8 percent of total imports (ranked first) by Korean firms; in 1992 alone, Korea’s trade deficit with Japan amounts $7.858 billion representing more than 1.5 times the total trade deficit

($5.144 billion) in 1992 [Koreas Foreign Trade Association 1993].

3. During the period of 1962-92, Japan’s investments in Korea represented 41.5 percent of entire foreigner’s investments in Korea [Korea Foreign Trade Association 1993].

4. Regression analysis using concurrent year’s independent variables provides similar results to those reported here with slightly lower R2 values.

5. The second adjusted measure is in line with Lucas (1993), who uses net investment of a country divided by that country’s deflator for fixed capital formation.

6. It is, however, plausible to have more FDI with a trade surplus since a surplus puts pressure on exchange rate and domestic prices on one hand and it makes acquisitions of funds for FDI easier on the other.

7. The efficiency-wage is measured as absolute nominal manufacturing wage differences divided by an index of relative labor productivity between Taiwan Province of China and each of the investing countries.

8. It is also plausible that Korea’s dependence on Japan leads to a high degree of transaction costs for Korea, which would defer their effort to look for alternative sources in response to the appreciation of Korean won over Japanese yen. Such dependence often creates bargaining problem as implied in the transaction cost economic perspective [see Rugman 1980]. Therefore, with small fluctuation in exchange rates, it would be more efficient to stay with Japan as the source of technologies and products.

9. The regression analysis.using adjusted KFDI and LJFDI by each country’s total annual investment as the dependent variable yielded results qualitatively similar to those reported in Table 3.

REFERENCES

Bandera, V.N. and J.T. White. 1968. “US Direct Investments and Domestic Markets in Europe.” Economia Internazianale (February), 117-133.

Hennart, J.F. and Y.R. Park. 1994. “Location, Governance, and Strategic Determinants of Japanese Manufacturing Investment in the United States.” Strategic Management Journal (July), 419-436.

Keizai Koho Center. Japan: An International Comparison. 1986-1994.

Kim, S.H. 1993. “Motives for Japanese Direct Investment in the United States.” Journal of

International Money dr Finance (October), 6672.

Komiya, R. 1991. “Japan’s Foreign Direct Investment.” Annals of the American Academy of Political & Social Science (January), 48-61.

Korea Foreign Trade Association. Korea and the World: Key Indicators. 1986-1994.

Korea Foreign Trade Association. Foreign Trade Trends and Key Indicators. 1986-1994.

Lall, S. 1980. “Monopolistic Advantages and Foreign Investment by US Manufacturing Industry.” Oxford Economic Papers (March), 102-122.

Lee, C.H. 1994. “Korea’s Direct Investment in Southeast Asia.” ASEAN Economic Bulletin (March), 280-296.

Lucas, R. 1993. “On the Determinants of Direct Foreign Investment: Evidence from East and Southeast Asia.” World Development (March), 391-406.

Lunn, J.L. 1980. “Determinants of U.S. Direct Investment in the EEC. European Economic Review (January), 93-101.

Mann, C.L. 1993. “Determinants of Japanese Direct Investment in U.S. Manufacturing Industries.” Journal of International Money & Finance (October), 523-541.

Matsuura, N.F. 1989. “Management Conflict and Foreign Direct Investment: The Case of Japanese Investment in South Korea.” Columbia Journal of World Business (Summer), 61-67.

Petrochilas, G.A. 1984. “The Determinants of Foreign Direct Investment in the Greek Economy.” British Review of Economic Issues (Spring), 27-54.

Reidel, J. 1975. “The Nature and Determinants of Export-Oriented Direct Foreign Investment in

a Developing Country: A Case Study of Taiwan.” Weltwirtschaftliches Archive, 505-528.

Rugman, A.M. 1980. “A New Theory of the Multinational Enterprise: Internationalization versus Internalization,” Columbia Journal of World Business, 23-29.

Scaperlanda, A. and R. Balough. 1983. “Determinants of U.S. Direct Investment in the EEC: Revisited.” European Economic Review (May), 381-390.

Scaperlanda, A. and L.J. Mauer. 1973. “The Impact of Controls on U.S. Direct Foreign Investment in the EEC.” Southern Economic Journal, 419-423.

Schmitz, A. and J. Bieri. 1970. “EEC Tariffs and U.S. Direct Investment.” Economic Journal, 724-731.

Serapio, M.G. Jr. 1993. “Macro-Micro Analysis of Japanese Direct R&D Investments in the U.S. Automotive and Electronic Industries.” Management International Review (Third Quarter), 209-225.

Torrisi, C.R. 1985. “The Determinants of Direct Foreign Investment in a Small LDC.” Journal of Economic Development (July), 29-45.

United Nations. 1992. The Determinants of Foreign Direct Investment: A Survey of the Evidence, United Nations publications.

The case of the Mercosur, The

Letter from the editor