GLOBALIZATION AND ECONOMIC GROWTH IN CAMBODIA

The Singapore Economic Review, Vol. 62, No. 2 (2017) 363 375 World Scientific Publishing Company DOI: 10.1142/S0217590815500708 GLOBALIZATION AND ECONOMIC GROWTH IN CAMBODIA JAI S. MAH Professor, Division of International Studies Ewha Womans University Seodaemun-gu, Seoul 120-750, South Korea jsmah@ewha.ac.kr Published 2 September 2015 For the past two decades, since the political situation became stable, Cambodia has recorded a very rapid economic growth rate. In the meantime, globalization has progressed both in terms of the expansion of international trade and foreign direct investment (FDI) inflows. This paper applies small sample cointegration tests and error correction models to reveal the determinants of Cambodia s rapid economic growth. The cointegration test results support the existence of a long-run equilibrium relationship among the variables concerned. The error-correction models show that expansion of international trade values has caused the rapid economic growth in Cambodia, regardless of the measure of international trade used. Keywords: Globalization; economic growth; Cambodia. JEL Classification: F21, F43 1. Introduction With the signing of the Paris Peace Agreements in 1991 and the free national election conducted by the United Nations in 1993, the political situation in Cambodia was stabilized. Since then, Cambodia has recorded a very rapid economic growth rate. Starting from a position as one of the poorest countries in the world, it has transformed itself to a country entering the status of a lower middle income developing country (National Institute of Statistics, 2012). It recorded an annual average real GDP growth rate of 7.7% during the period 1993 2013. Due to this rapid growth, Cambodia s GDP per capita rose from US$251 in 1993 to US$1,025 in 2013 (Asian Development Bank (ADB), 2014; World Bank, 2014). Globalization has proceeded rapidly since 1993. The annual average trade dependence ratio defined as (export values þ import values)/gdp was recorded as 48.7% in 1993 and it continued to increase to over 121% during 2010 2013. For international trade, Cambodia continued to record trade deficits for the whole period under consideration. The majority of export products were garments. The higher economic growth rates were accompanied by a large amount of foreign direct investment (FDI) inflows in the same period. That is, 363

364 The Singapore Economic Review although the share of FDI inflows in GDP was lower than 2.5% in 1993 and 1994, it was recorded as being over 7% during 2010 2013. Despite remarkable economic growth in Cambodia for the past two decades, to the best of the author s knowledge, no rigorous empirical work has thus far revealed the causes. This paper reveals the determinants of economic growth in Cambodia by using time series data for the period 1993 2013 and focuses on the impact of globalization in the sense that its progress has been noteworthy. Since this paper employs time series data, it starts by applying unit root tests. Considering the limited number of observations in the data set, it employs a small sample cointegration test. Since this cointegration test shows that all the variables concerned are cointegrated, the error correction model is used and its estimation results show that the trade dependence ratio causes economic growth in Cambodia. Neither investment nor FDI is revealed to cause economic growth. The structure of this paper is as follows: Section 2 provides the overview of globalization and economic growth in Cambodia for the past two decades. Sections 3 and 4 address the relevant literature and the model for the empirical work, respectively. Section 5 shows the empirical evidence drawn from small sample cointegration tests and error correction models to test causality. Conclusions are provided in Section 6. 2. Overview of Globalization and Economic Growth in Cambodia, 1993 2013 Although Cambodia is still a low-income developing country, it has recorded very rapid economic growth for the past two decades. As Table 1 shows, the annual average real GDP growth rate rose throughout the 1990s and the early to mid-2000s, reaching 10.6% during 2003 2007. Despite its slowdown to 5.4% during 2008 2012, it rose again to 7.5% in 2013. The size of the population increased from less than 10 million in the late 1980s to 15 million in 2013. Notwithstanding this population growth, the unemployment rate has been less than 3% throughout the period except for 1998, when it was 5.3%. The inflation rate remained in single digit throughout the 1990s and 2000s with only a few exceptions. Together with macroeconomic stability, globalization has proceeded rapidly in the Cambodian economy (ADB, 2014). As a relatively small country, Cambodia has tried to pursue an outward-oriented economic development strategy since the early 1990s. As shown in Table 1, the trade Table 1. Overview of Globalization and Economic Growth Rate (unit: annual average, %) Years Real GDP Growth Rate Trade Dependence Ratio Export/GDP Ratio FDI/GDP Ratio 1993 1997 6.6 67.8 26.4 4.7 1998 2002 8.1 103.0 45.9 5.1 2003 2007 10.6 135.5 63.6 5.4 2008 2012 5.4 115.8 55.5 7.6 2013 7.5 145.3 66.1 8.5 Source: ADB, Key Indicators for Asia and the Pacific, retrieved September 8, 2014.

Globalization and Economic Growth in Cambodia 365 Table 2. Exports, Imports, Trade Balance and FDI Inflows (unit: US $ billion) Year Exports Imports Trade Balance FDI Inflows (US $ Million) 1993 0.3 0.5 0.2 54 1998 0.8 1.2 0.4 223 2003 2.1 2.7 0.6 74 2008 3.5 5.1 1.6 795 2013 6.5 9.5 3.0 1,299 Source: ADB, Key Indicators for Asia and the Pacific, retrieved September 8, 2014. dependence ratio rose from 67.8% during 1993 1997 to 145.3% in 2013, which is one of the highest in the world except for a few city states. The exports/gdp ratio continued to rise from less than 30% during 1993 1997 to 66.1% in 2013. Together with the rapid increase in the values of international trade, the increase in the amount of FDI inflows has been remarkable as well. That is, the ratio of FDI/GDP continued to rise from 4.7% during 1993 1997 to 8.5% in 2013. During the period of rapid economic growth and progress of globalization, the Cambodian economy has recorded trade deficits. As Table 2 shows, export values increased from less than US$.3 billion in 1993 to US$6.5 billion in 2013. In the meantime, import values also increased from US$.5 billion in 1993 to US$9.5 billion in 2013. Consequently, the trade deficit increased from US$.2 billion in 1993 to US$3.0 billion in 2013. The amount of FDI inflows also rose from a mere US$54 million in 1993 to about US$1.3 billion in 2013. According to the WTO, during the 2000s, the three main sectors, garments, construction and tourism, represented more than 80% of total FDI inflows. China is Cambodia s largest investor (WTO, 2011: 10). In terms of the direction of international trade, Cambodia s exports are mainly directed to developed countries. For instance, in 2013, the United States shared 28.2% of Cambodia s exports. This was followed by the United Kingdom (8.7%) and Germany (8.4%). Unlike exports, Cambodia imports mainly from its neighboring, somewhat more developed countries. For instance, in the same year, Thailand and China shared 27.0% and 22.0% of Cambodia s imports, respectively. Vietnam accounted for 19.3% of Cambodia s imports (ADB, 2014). By product, garment exports accounted for between 80 and 90% of total exports in the late 2000s. For the same period, manufacturing, including textiles, machinery, and transport equipment accounted for 80% of imports (World Trade Organization (WTO), 2011: 10). As Table 3 shows, Cambodia s GDP per capita rose from US$267 in 1993 to US$350 in 2003, to US$742 and then to US$1,025 in 2013. Cambodia s economic growth has been accompanied by industrialization and an increase in domestic investment. For instance, the share of the agricultural sector relative to GDP fell from 46.4% in 1993 to 33.8% in 2013. On the other hand, that of the manufacturing sector rose from 8.9% in 1993 to 18.2% in 2003. Since then, it fell somewhat to 15.3% in 2013. The agricultural

366 The Singapore Economic Review Table 3. GDP Per Capita and the Industrial Structure (unit: %) Year GDP per Capita (US $) Agriculture in GDP Manufacturing Sector in GDP Gross Domestic Capital Formation in GDP 1993 267 46.4 8.9 11.0 1998 258 46.3 12.7 11.8 2003 350 33.6 18.2 22.0 2008 742 34.9 15.3 18.6 2013 1,025 33.8 15.3 16.2 Source: ADB, Key Indicators for Asia and the Pacific, retrieved September 8, 2014. sector still plays a very important role in the Cambodian economy in the sense that it employed as much as 64% of the labor force in 2013. The manufacturing sector employed 8% of the labor force in the same year. Although the ratio of gross domestic capital formation was as low as 11.0% in 1993, it rose rapidly to 22.0% in 2003. In 2013, it was recorded as 16.2%. As is the case of the other developing countries, insufficient infrastructure can be regarded as a bottleneck in the process of continuing economic growth. For instance, electricity production increased from 180 million kwh in 1993 to 540 million kwh in 2003 and then to 1.0 billion kwh in 2011 (ADB, 2014). Meanwhile, due to the rapidly increasing need for electricity as a result of rapid economic growth and industrialization, many companies still complain about the insufficient supply of electricity and consequent high electricity price. 1 Similarly, a lack of highways, railroads and well-maintained paved roads leads to high transportation costs. 3. Review of the Literature The typical production function in economic analysis shows that the investment, among other factors, may lead to economic growth. Covering a cross-section of 96 countries, Hall et al. (2010) ordinary least squares (OLS) estimation results showed that in countries with strong institutions, increases in human and physical capital have a greater effect on economic growth rates than in countries with bad institutions. Using China s provincial data for the period 1997 2009, Jarreau and Poncet s (2012) cross-sectional estimation results showed that physical capital accumulation has an insignificant effect on economic growth. Mehic et al. (2013) employed panel data for seven southeast European transition economies during 1998 2007 and show that domestic investment does not have a significant effect on economic growth. Besides investment as a whole, FDI inflows have often attracted the development economists attention. In addition to the aspect of the generation of job opportunities, they 1 Source: conversation with the Cambodian government officials in Phnom Penh on August 22, 2013.

Globalization and Economic Growth in Cambodia 367 may be beneficial to the economic growth of developing countries through the transfer of advanced technologies. However, they may affect the economic growth of the host country negatively in the case of a substantial amount of repatriation of profits and dividends, a large amount of royalty payments, or foreign invested enterprises (FIEs) crowding out indigenous enterprises (see, for instance, Zhang, 2001; Ram and Zhang, 2002; Ramirez, 2006). Although many empirical works have examined the effect of FDI inflows on economic growth, the conclusions have been mixed. Zhang (2001) tested the causality between these two variables by using real FDI stock and real GDP for 11 developing countries in East Asia and Latin America. His results were based on the Johansen cointegration test, the error correction model and the Granger causality test and showed that the impact of FDI on host countries depends on the situation of each country. Alguacil et al. (2002) applied the Granger causality test to data for Mexico and their results supported the FDI-led economic growth relationship. Choe (2003) tested the causality among economic growth, FDI inflows and gross domestic investment (GDI) in 80 countries. According to his study, there is bidirectional causality between economic growth and FDI inflows, although there is no causality from GDI to economic growth. Akinlo (2004) analyzed the case of Nigeria, an economy in which extractive industries, not manufacturing industries, take the lion s share of FDI inflows. His results, based on the Johansen cointegration test and the error correction model, showed that extractive FDI, crude oil in particular, might not lead to economic growth as much as manufacturing FDI, while exports, labor and human capital are positively related to economic growth. Ramirez (2006) applied the unit root test and the Johansen cointegration test to Mexico. His error correction model estimation results showed that the growth rate of the foreign capital stock per worker has a positive and significant effect on the labor productivity growth rate. Khamfula s (2007) regression results based on data for nine countries showed that the positive effect of FDI inflows on economic growth tends to become greater when a country becomes less corrupt. Duttaray et al. (2008) selected 66 developing countries and used time series data for 1970 1996 in relation to four variables, i.e., the real GDP growth rate, exports/gdp, FDI inflows/gdp, and productivity defined as the ratio of real GDP to total labor. Their Granger causality test results showed that FDI caused economic growth in 29 countries, but for the majority of developing countries, economic growth did not cause FDI at all. This means that the causality from FDI inflows to economic growth appears to be country specific. Mah (2010) applied a small sample cointegration test to the case of South Korea. His results showed that a changing degree of trade openness does not influence the impact of FDI inflows on economic growth. Mehic et al. (2013) showed that FDI had a significant positive effect on real GDP per capita in southeast European economies during the period 1998 2007. The conclusions of the papers mentioned above appear to be mixed with respect to the effect of FDI inflows on economic growth. Together with the literature dealing with the effect of FDI inflows on economic growth, there have been many empirical works focusing on the effect of international trade on economic growth in developing countries. The hypothesis that export growth leads to economic growth, i.e., the export-led growth hypothesis, has received considerable

368 The Singapore Economic Review attention from economists. According to this hypothesis, an expansion in exports leads to economic growth due to the positive externality, the economies of scale associated with expanded sales opportunities, and the improvement of the technology level due to the harsher competition that exporters face. Richards (2001) performed the Granger causality test and Engle-Granger two-step cointegration test with respect to economic growth, export growth, investment/gdp, the growth of the labor force and cultivated land size of Paraguay. The empirical evidence showed that economic growth has a positive long-run effect on the export share of national products, whereas the causality from the latter to the former is not supported. Krishina et al. (2003) used panel data for 39 developing countries covering the period 1951 1998. Exports were revealed to be much more important in GDP growth regressions involving countries with GDP growth rates below the mean panel GDP growth rate than in those with GDP growth rates above the panel mean. Their results suggest that analysis at the individual country level may provide useful information for uncovering the determinants of economic growth. Cuaresma and Worz (2005) used a data set covering 45 developed and developing countries and tested the hypothesis that exports of technology-intensive products have a higher potential for positive externalities coupled with higher productivity levels. Their estimation results support the superior performance of high-tech exports. Using data for some 20 developing countries, Kneller (2007) showed that countries liberalizing their trade regime tend to increase their spending on social welfare. However, even controlling for the effect of fiscal policy, trade liberalization did not cause changes in economic growth. Kim et la. (2009) used data for South Korea covering the period 1980 2003. The Granger causality test indicated that exports had no significant positive effect on either productivity or GDP growth, providing little support for the export-led growth hypothesis. Thus, the conclusions of many empirical works since the 2000s using time series methods do not unanimously support the trade (export)-led economic growth hypothesis. 4. The Model The economic growth model starts from a typical production function. In developing countries that rely heavily on the agricultural sector, due to the high rate of disguised unemployment in the agricultural sector, population growth can be assumed not to lead to economic growth, whereas a change in capital formation is expected to result in economic growth. The expansion of international trade and FDI inflows may raise the economic growth rate via their contribution to the improvement of the technology level and/or enhancing capital stock. Therefore, in this paper, economic growth is captured by the following equation: YðtÞ ¼a 0 þ a 1 INVðtÞþa 2 FDIðtÞþa 3 TDRðtÞþeðtÞ, ð1þ where Y, INV, FDI and TDR denote the real GDP growth rate, gross capital formation/ GDP, FDI inflows/gdp, and trade dependence ratio defined as (export values þ import values)/gdp, respectively. The paper also examines the results when the export

Globalization and Economic Growth in Cambodia 369 dependence ratio (EDR) and import dependence ratio (IDR), defined as export values/gdp and import values/gdp, respectively, are used instead of TDR. The annually observed data cover the period 1993 2013 and are drawn from the ADB (2014) and the World Bank (2014). FIEs, most of which are related to garment production, account for the majority of investment in Cambodia. For instance, during the period 1994 2010, approximately 70% of total investment was born by FIEs, including wholly foreign-owned firms and joint ventures (National Institute of Statistics, 2012). Thus, FDI in Equation (1) may represent the investment activities in general. Reflecting this, economic growth in Cambodia is also examined by Equation (2): YðtÞ ¼b 0 þ b 1 FDIðtÞþb 2 TDRðtÞþeðtÞ: ð2þ 5. Empirical Evidence As this paper uses time series data, the augmented Dickey Fuller and Phillips Perron unit root tests are applied to the variables to test their stationarity. Optimal lags are selected using the Schwarz lag selection criterion for the former and the Newey West method for the latter. The results of both the augmented Dickey Fuller and Phillips Perron unit root tests show that all variables concerned are integrated of order one at the 5 or 10% level of significance, which is available upon request. If the variables under consideration are non-stationary, it is necessary to examine whether or not there are long-run equilibrium relationships among the variables concerned, using cointegration tests. As the data period is short, i.e., 1993 2013, the study employs Banerjee et al. (1998) small sample cointegration test, which can be applied to nonstationary variables with the same order of integration. It depends on the statistical significance of the estimated OLS coefficient of the lagged dependent variable in an autoregressive distributed lag model augmented with leads of the regressors. The cointegration test is based on an equation estimated as follows: AðLÞdYðtÞ ¼BðLÞdXðtÞþbYðt 1ÞþCXðt 1Þþ s i¼1f i dxðt þ iþþuðtþ, ð3þ where Y, X and u show the regress and, the vector of regressors, and the conventionally assumed error term, respectively. L and d denote the lag operator and the differenced form of the concerned variable, respectively. The cointegration test results depend on the estimated t statistics of the parameter b in Equation (3). Under the null hypothesis of noncointegration, b ¼ 0; under the alternative hypothesis of cointegration, 2 < b < 0. Banerjee et al. (1998) shows that their statistics have power in finite samples with a limited number of observations that is greater than the other cointegration tests more frequently used. Table 4 shows Banerjee, Dolado, and Mestre s small sample cointegration test results when INV, FDI and TDR are used as the right-hand side variables in Equation (1). It also shows the cointegration test results when FDI and EDR or FDI and IDR are used as the right-hand side variables. Regardless of the number of leads of s in Equation (2), Table 4

370 The Singapore Economic Review Table 4. Banerjee, Dolado, and Mestre s Small Sample Cointegration Test Results Variables Lead Number ¼ 1 Lead Number ¼ 2 Y, INV, FDI, TDR 4.387** 10.000** Y, FDI, TDR 4.573** 3.320* Y, FDI, EDR 5.159** 3.341* Y, FDI, IDR 5.026** 3.044* *Statistically significant at the 5% level of significance. **Statistically significant at the 1% level of significance. shows that the null hypothesis of non-cointegration is rejected at the 1 or 5% level of significance. That is, the small sample cointegration test shows that the variables concerned are cointegrated. Table 5 shows the estimation results of the error correction model based on Equation (1). When Y is used as the left-hand side variable, the estimated coefficient of the error correction term is negative and statistically significant at the 1% level. The estimated coefficient of TDR is positive and statistically significant at the 1% level. That is, TDR is Table 5. Estimation Results of the Error Correction Model Applied to Equation (1) Left-Hand Side Variable Right-Hand Side Variable Y INV FDI TDR Constant 1.031 0.045 0.226 1.214 ( 2.277) (0.140) (0.335) (1.342) EC(t-1) 1.113** 2.646** 0.605 0.604 ( 3.647) ( 4.818) ( 1.326) ( 2.559) INV(t) 0.144 0.728 1.493 ( 0.587) ( 1.727) (2.267) FDI(t) 0.233 0.366* 0.955 (1.175) ( 2.316) (1.473) TDR(t) 0.268** 0.150* 0.262 (5.964) (2.800) (0.952) YðtÞ 0.278 0.283 1.563* ( 1.750) ( 0.425) (3.796) Adjusted R 2 0.780 0.706 0.113 0.936 Durbin Watson statistic 1.822 2.200 2.250 2.186 F-statistic 8.107** 5.804* 1.160 19.392* Notes: Values within the parentheses denote the calculated t-statistics. The estimated coefficients of the lagged terms in the right-hand side of the error correction model are not reported in this table. *Statistically significant at the 5% level of significance. **Statistically significant at the 1% level of significance.

Globalization and Economic Growth in Cambodia 371 shown to cause Y. Meanwhile, FDI is not statistically significant at any reasonable level of significance. When TDR or FDI is used as the left-hand side variable, the error correction term is not significant. In the case that INV is used as the left-hand side variable, the error correction term is negative and statistically significant at the 1% level. In that case, the estimated coefficients of FDI and TDR are negative and positive, respectively, at the 5% level of significance. That is, it appears that a higher trade dependence ratio leads to a higher investment ratio, while FDI inflows tend to crowd out domestic investment. The crowding-out of domestic investment by FDI on economic growth may be partially responsible for the insignificant impact of FDI inflows on economic growth. Using data for the Central and Eastern European countries, Mehic et al. (2013) also reported some degree of substitutability between FDI and domestic investment. The estimation results for the error correction model based on Equation (2) are reported in Table 6. Although INV is omitted from the error correction model estimated, the overall results do not change qualitatively. First, when Y is used as the left-hand side variable, the estimated model appears to be plausible in terms of the explanatory power. The estimated coefficient of the error correction term is 0.816 and statistically significant at the 5% level, confirming the result of the cointegration test. TDR is positive and statistically significant at the 1% level. That is, TDR is shown to cause economic growth. Although the estimated coefficient of FDI is positive, it is not statistically significant at the 5% level. Table 6. Estimation Results of the Error Correction Model Applied to Equation (2) Left-Hand Side Variable Right-Hand Side Variables Y FDI TDR Constant 0.879 0.311 3.670 ( 1.514) (0.506) (1.394) EC(t-1) 0.816* 0.267 0.089 ( 2.592) ( 0.862) ( 0.668) FDI(t) 0.479 1.900 (2.120) ( 1.592) TDR(t) 0.164** 0.094 (3.794) ( 1.565) YðtÞ 0.395 2.618* (1.573) (3.002) Adjusted R 2 0.605 0.119 0.393 Durbin Watson statistic 1.832 2.224 1.873 F-statistic 5.331** 1.382 2.831 Notes: Values within the parentheses denote the calculated t-statistics. The estimated coefficients of the lagged terms in the right-hand side of the error correction model are not reported in this table. *Statistically significant at the 5% level of significance. **Statistically significant at the 1% level of significance.

372 The Singapore Economic Review Table 7. Estimation Results of the Error Correction Model when EDR is used as the Measure of International Trade in Equation (2) Left-Hand Side Variable Right-Hand Side Variables Y FDI EDR Constant 1.067 0.340 2.305 ( 1.864) (0.531) (1.748) EC(t-1) 0.858** 0.334 0.117 ( 3.150) ( 1.119) ( 1.071) FDI(t) 0.495* 0.838 (2.443) ( 1.448) EDR(t) 0.355** 0.181 (4.297) ( 1.527) YðtÞ 0.376 1.209* (1.530) (2.715) Adjusted R 2 0.661 0.117 0.442 Durbin Watson statistic 1.914 2.301 1.954 F-statistic 6.533** 1.375 3.244* Notes: Values within the parentheses denote the calculated t-statistics. The estimated coefficients of the lagged terms in the right-hand side of the error correction model are not reported in this table. *Statistically significant at the 5% level of significance. **Statistically significant at the 1% level of significance. Table 6 also shows the results of the error correction models when FDI or TDR is used as the left-hand side variable. The error correction term is statistically insignificant at any reasonable level of significance in such cases. The estimation results for the error correction model when EDR is used instead of TDR as the measure of international trade are reported in Table 7. The coefficient of the error correction term is found to be 0.858 and statistically significant at the 1% level, confirming the result of the cointegration test. The coefficient of EDR is estimated to be positive and statistically significant at the 1% level. This means that export expansion causes economic growth in Cambodia. FDI is revealed to be statistically significant at the 5% level. Table 7 also shows the results of the error-correction models when FDI or EDR is used as the left-hand side variable. The error correction term is statistically not significant at the 5% level in such cases. None of these variables entered on the right-hand side of the error correction model is statistically significant at the 5 or 1% level. Such results imply that EDR (FDI) is not caused by Y or FDI (EDR). As the effect of the expansion of import values on economic growth may be entirely different from that of export values or trade values as a whole, Table 8 reports the estimation results for the error correction model when IDR is used as the measure of international trade. The coefficient of the error correction term is estimated as 0.925 and statistically significant at the 5% level, that of IDR is shown to be positive and statistically significant at the 5% level. This means that import expansion causes economic growth in

Globalization and Economic Growth in Cambodia 373 Table 8. Estimation Results of the Error Correction Model when IDR is used as the Measure of International Trade in Equation (2) Left-Hand Side Variable Right-Hand Side Variable Y FDI IDR Constant 0.747 0.270 1.657 ( 1.235) (0.441) (1.119) EC(t-1) 0.925* 0.250 0.101 ( 2.550) ( 0.752) ( 0.552) FDI(t) 0.410 0.953 (1.595) ( 1.377) IDR(t) 0.261* 0.146 (2.886) ( 1.287) YðtÞ 0.324 1.284* (1.301) (2.667) Adjusted R 2 0.510 0.048 0.256 Durbin Watson statistic 1.830 2.057 1.595 F-statistic 3.955* 1.144 1.973 Note: Values within the parentheses denote the calculated t-statistics. The estimated coefficients of the lagged terms in the right-hand side of the error correction model are not reported in this table. *Statistically significant at the 5% level of significance. **Statistically significant at the 1% level of significance. Cambodia, which is qualitatively the same as the effect of export expansion on economic growth. FDI is statistically insignificant at the 5% level, which confirms the result in Table 6. Table 8 also shows the estimation results when FDI or IDR is used as the left-hand side variable. In such cases, the error correction term is revealed to be statistically insignificant at any reasonable level of significance. Therefore, we can conclude that there is no evidence that FDI (IDR) is caused by Y or IDR (FDI). The results of the small sample cointegration tests and error correction models show that the real GDP growth rate, investment, FDI inflows, and the trade dependence ratio have a long-run equilibrium relationship in Cambodia. The error correction models show that the trade dependence ratio causes economic growth regardless of the measure of international trade, whereas there is no significant evidence that FDI inflows or investment cause economic growth in Cambodia. The finding that investment does not cause economic growth seems to be consistent with the findings of many authors such as Richards (2001), Choe (2003) and Mehic et al. (2013). For instance, Richards (2001) Granger causality test and Engle, Granger two-step cointegration test show that the effect of investment on economic growth is not statistically significant in Paraguay. According to Choe s (2003) study using panel data on 80 countries over the period 1975 1995, GDI does not cause economic growth. Mehic et al. (2013) research employing data on seven European

374 The Singapore Economic Review transition economies shows that domestic investment has no significant effect on growth, either. As suggested by Hall et al. (2010), in a country such as Cambodia where institutions are weak, investment may not have a large effect on economic growth. 6. Conclusion For the past two decades, since the political and social situation became stable, Cambodia has recorded a very high economic growth rate. In the meantime, globalization in terms of both the expansion of trade values and FDI inflows has progressed. This paper applies Banerjee et al. (1998) small sample cointegration tests and error correction models to reveal the causes of economic growth. The cointegration test results support the existence of a long-run equilibrium relationship among the variables concerned. The results of the error correction models show that the rise in the trade dependence ratio has caused rapid economic growth in Cambodia. It is intuitively plausible in the sense that since the mid- 1990s, the trade dependence ratio has been very high. Meanwhile, there is little evidence of FDI inflows causing economic growth. Domestic investment is revealed not to cause economic growth in Cambodia. Although the paper uses the EDR and the IDR in addition to the trade dependence ratio, the overall results do not change regardless of the measure of international trade employed. That is, both export expansion and import expansion are shown to cause economic growth of Cambodia. Such results provide important policy implications in the sense that pursuing both export promotion policies and import liberalization measures, allowing freer imports of intermediate goods and capital goods, would contribute to economic growth. However, despite the findings reported here, the results should be treated cautiously due to the short period of analysis. References Akinlo, AE (2004). Foreign Direct Investment and Growth in Nigeria: An Empirical Investigation. Journal of Policy Modeling, 26(5), 627 639. Alguacil, M, TA Cuadros and V Orts (2002). Foreign Direct Investment, Exports and Domestic Performance in Mexico: A Causality Analysis. Economics Letters, 77(3) 371 376. Asian Development Bank, Key Indicators for Asia and the Pacific, available at http://www.adb. org, retrieved September 8, 2014. Banerjee, A, JJ Dolado and R Mestre (1998). Error-Correction Mechanism Tests for Cointegration in a Single-Equation Framework. Journal of Time Series Analysis 19(3), 267 283. Choe, J Il (2003). Do Foreign Direct Investment and Gross Domestic Investment Promote Economic Growth? Review of Development Economics, 7(1), 44 57. Cuaresma, JC and J Worz (2005). On Export Composition and Growth. Review of World Economics, 141(1), 33 49. Duttaray, M, AK Dutt and K Mukhopadhyay (2008). Foreign Direct Investment and Economic Growth in Less Developed Countries: An Empirical Study of Causality and Mechanisms. Applied Economics, 40(15), 1927 1939. Hall, JC, RS Sobel and GR Crowley (2010). Institutions, Capital, and Growth. Southern Economic Journal, 77(2), 385 405.

Globalization and Economic Growth in Cambodia 375 Jarreau, J and S Poncet (2012). Export Sophistication and Economic Growth: Evidence from China. Journal of Development Economics, 97, 281 292. Khamfula, Y (2007). Foreign Direct Investment and Economic Growth in EP and IS Countries: The Role of Corruption. The World Economy, 30(12), 1843 1854. Kim, S, HJ Lim and D Park (2009). Imports, Exports and Total Factor Productivity in Korea. Applied Economics, 41(14), 1819 1834. Kneller, R (2007). No Miracles Here: Trade Policy, Fiscal Policy and Economic Growth. Journal of Development Studies, 43(7), 1248 1269. Krishina, K, A Ozyildirim and NR Swanson (2003). Trade, Investment and Growth: Nexus, Analysis and Prognosis. Journal of Development Economics, 70(2), 479 499. Mah, JS (2010). Foreign Direct Investment Inflows and Economic Growth: The Case of Korea. Review of Development Economics, 14(4), 726 735. Mehic, E, S Silajdzic and V Babic-Hodavic (2013). The Impact of FDI on Economic Growth: Some Evidence from Southeast Europe. Emerging Markets Finance and Trade, 49, 5 20. National Institute of Statistics (NIS), Statistical Yearbook of Cambodia 2011, NIS: Phnom Penh, 2012. Ram, R and KH Zhang (2002). Foreign Direct Investment and Economic Growth: Evidence from Cross-country Data for the 1990s. Economic Development and Cultural Change, 51(1), 205 215. Ramirez, MD (2006). Is Foreign Direct Investment Beneficial for Mexico? An Empirical Analysis, 1960 2001. World Development, 34(5), 802 817. Richards, DG (2001). Exports as a Determinant of Long-Run Growth in Paraguay, 1966 1996. Journal of Development Studies, 38(1), 128 146. World Bank, Available at: http://databank.worldbank.org/data, retrieved September 7, 2014. World Trade Organization (WTO), Trade Policy Review Cambodia, WTO: Geneva, 2011. Zhang, KH (2001). Does Foreign Direct Investment Promote Economic Growth? Evidence from East Asia and Latin America. Contemporary Economic Policy, 19(2), 175 185.