In this chapter, we will discuss the background of foreign direct investment (FDI) among ASEAN countries, background of FDI for selected ASEAN countries (Indonesia, Malaysia and Thailand), problem statement, objectives, significant of study and the layout of the study.
1.1 The Background of Foreign Direct Investment (FDI) among ASEAN
In 1967, ASEAN is established with the alliance of five countries (Malaysia, Thailand, Indonesia, Singapore, and Philippines). Bangkok declaration has been signed and agreed on the 5 members in ASEAN. ASEAN has expanded including Brunei, Vietnam, Laos, Burma and Cambodia in 1984, 1995, 1997 and 1999. All together ASEAN has ten member countries.
Regional differences have been recognized as significant in contribution to the increasing trend of FDI. The emerging regions of ASEAN and Latin America have achieved strong growth in FDI inflows. According to World Investment Report (2011), some of the poorest regions continue to show decreasing trend in FDI inflows such as Least Developed Countries (LDCs), Landlocked Developing Countries (LLDCs) and Small Island developing States (SIDS).
FDI has contributed to the country development and gross fixed capital formation in developing countries (Hill, 1994). Also, FDI plays a significant role of fundamental factor in a country economic developmental process (Jensen, 2006). Therefore, FDI has become significantly important in the developing countries (Lipsey & Sjoholm, 2011). Anwar and Nguyen (2010) have observed that an increased in FDI will bring positive impact toward GDP growth and has contributed to economic growth in developing countries.
There are a number of developing countries that have succeeded in attracting substantial FDI inflows and the growth of economic transformation (Srinivasan, Kalaivani, & Ibrahim, 2010). The increase in FDI reflects the interest and confidence of investors toward the investment or operating business in the region (Karimi, Yusop, & Law, 2010). It leads to the strengthening of the region's economy. Hence, developing countries are concerned about the consistent growth of FDI in order to enhance the performance of economic and the ability to compete with other regions. Furthermore, ASEAN plays a significant role in international trade especially in Intra-Asia trade. Intra-Asia trade shows an upward trend as the duties on hundred products had been removed. Moreover, FDI inflows is rising as ASEAN address the Free Trade Agreement (FTA) that focuses on the economic and technical cooperation in areas such as energy, information technology, and industrial cooperation (Sen & Srivastava, 2009).
Table 1.1: Net Inflows of FDI among AEAN in 2010
Source: World Bank
Table 1.1 shows the net inflows of FDI in ASEAN in 2010. Singapore has achieved the highest FDI as compared to other members in ASEAN. This is due to the fact that Singapore is a developed country where it has a stable economic and political condition. Besides, higher levels of infrastructure technology and professional supply have attracted more FDI inflows to Singapore. The second highest of FDI net inflows is Indonesia with US$13,400,000,000. Indonesia economy has concentrated on the development of oil and gas sector as an oil producing country. Thus, the oil and gas sector plays an important role in supporting the upward trend of FDI inflows in Indonesia. Furthermore, Thailand is ranked the third highest FDI with US$9,678,888,214 net inflows in 2010. This is due to the outperforming of economic growth in the region. Malaysia is ranked the fourth place among the ASEAN countries with US$9,167,201,907 in 2010. The World Bank Economic Monitor of Malaysia has observed that the FDI increase in 2010. This is because of the achievement of Malaysia's Economic Transformation Program which is proposed in the New Economic Model that particularly in raising awareness on the availability of ready-to investment projects. Vietnam is ranked the fifth place with US$8,000,000,000. Vietnam has gained ground as low-cost production locations, especially for low-end manufacturing has boosted up the level of FDI inflows.
The level of net inflows of FDI after Vietnam is follow by Philippines, Myanmar, Cambodia, Brunei and Laos accordingly. Philippines has US$1,713,000,000 focusing on chemical and healthcare products, electronics, air-conditioning system, electricity, gas, water, financial intermediation, real estate, mining, and construction industries. Moreover, Myanmar has obtained US$910,345,755 with extractive industries while Cambodia has achieved US$782,596,735 with the main FDI inflows in agricultures, garments, banks and telecommunications. Besides that, Brunei has achieved US$495,720,000 in net inflows of FDI because European Union (EU) is the Brunei Darussalam's largest source of FDI. Lastly, Laos has the lowest FDI net inflow with US$278,805,903. Laos is recognized as least developed country that involved in agriculture, finance and manufacturing.
Table 1.2: Net Inflows of FDI in Selected Countries among ASEAN in 2010
Source: World Bank
Table 1.2 shows increasing trends of the FDI net inflows in Thailand, Malaysia and Indonesia in 1981-2010. These three countries are the focus in this study as they are the largest FDI recipients among other ASEAN countries Together they account for more than 90% of flows to the sub-region in 1997-2007 (Srinivasan et al., 2010).
Because this study is focusing on the developing countries, therefore Singapore is excluded as it is already a developed country although it has the highest FDI inflows among all ASEAN countries.
1.2 Background of FDI for Selected ASEAN Countries
1.2.1 Malaysia
Over the past decade, Malaysia has experienced a substantial growth in their economic, especially in the form of FDI. Malaysia is one of the countries that achieved the highest FDI amongst Thailand, Indonesia and Philippines (Mithani, Ahmad & Saifudin, 2008) and has the strategic demographic advantage of tangible assets like natural resources to attract the FDI. In the past decade, Malaysia FDI inflows have been increasing dramatically. In 1986, the enforcement of the Investment Act has improved trade liberalization by reducing the restrictions over capital ownership of foreign companies and has driven vast FDI inflows into Malaysia (Jajri, 2009). FDI inflows have affected the economic growth through its impact on capital stock, technology transfer, skill acquisition, or market competition (Aurangzeb & Haq, 2012). In addition, Malaysia has benefited from FDI through structural transformation and has pushed the country economic to higher stages (Athukorala & Wagle, 2011). These can be seen in Table 1.3 where the increasing trend of FDI inflows of Malaysia has led to the increased in GDP in Malaysia.
However, the trend of FDI inflows has dropped sharply from US$3.9 billion in 1999 to US$554 million in 2001 and US$8.6 billion in 2007 to US$1.4 billion in 2009. This sharp drop is due to the Asian financial crisis in 1997-1998 (Jeon, 2010) and global financial crisis that has happened in 2008 (Morris, Pham & Gray, 2011). Jajri (2009) explain that the Asian financial crisis in 1997 has affected most of the panic investor's decision to pull out short-term capital on a large scale which directly reduces the FDI inflows into Malaysia. After the financial crisis in 1997, the FDI inflows has regained and achieving its peak in 2007 which amounted US$8.6 billion.
Besides that, the trend of Malaysia FDI inflows also shows the fluctuating movement due to the volatility of exchange rate. Brzozowski (2006) states that exchange rate volatility has significant impact on FDI. Table 1.3 shows both trend of FDI inflows and exchange rate of Malaysia has same characteristic of fluctuation movements. The FDI pattern can be in a number of ways where appreciation and depreciation of the local currency is affected by the increase in exchange rate volatility. In normal conditions, potential investors will invest in a foreign country as long as the expected returns are high enough to cover currency risk (Hau & Rey, 2006; Pesenti & Tille, 2000). When the exchange rate volatility is high means there is a higher investment risk. Thus, FDI will be lower when exchange rate volatility is higher (Chowdhury & Wheeler, 2008).
Table 1.3: GDP, FDI Inflows and Real Exchange Rate of Malaysia in 1990-2010
Gross domestic product (GDP) (US$)
FDI inflows
Real exchange rate, index
Source: World Bank
1.2.2 Indonesia
Indonesia has vast natural resources, huge potential of domestic market, competitive and productive labour force, and a market-oriented economic policy that highly attracting FDI inflows. In 1960 and 1970, oil and gas is the main focus sector that is able to attract huge amount of FDI in Indonesia. Since 1980 Indonesia government has set a main economic policy to encourage FDI in the broad-based manufacturing sector and to become invigorate in export-oriented country. It has denominated export in rubber and petroleum where petroleum is one of the natural recourses that are drawing attention from other regions. Besides, Indonesia has implemented some economic policies, for example, the targeted nature of expenditure cuts and the employment opportunities (Hill, 1994; Warnecke & Ruyter, 2012). Furthermore, labor intensive manufacturing export industries in Indonesia are becoming more significant in sustaining the country's economic growth momentum (Posso, 2011; Coxhead & Li, 2008; Narjoko & Jotzo, 2007). This allows Indonesia to enjoy impressive economic growth rate in past decade. Table 1.4 shows the increasing trend in FDI inflows of Indonesia.
The trend of FDI inflows has dropped drastically from US$6.2 billion in 1996 to negative amount which is US$4.6 billion in 2000 and remained negative until 2003. FDI has significantly dropped due to the Asian financial crisis that happened in 1997-1998 (Aswicahyono, Hill & Narjoko, 2010; Miyakoshi, Okubo & Shimada, 2009; Sufian & Habibullah, 2010). The arrival of the Asian financial crisis in 1997 left a huge impact in Indonesia's financial sector, public institutions, and corporate governance. As a result of that, it brings adverse effect on FDI, trade, and finance and has driven Indonesia's economy to drop (Rajenthran, 2002). However, FDI inflows of Indonesia has recovered with a consistant increasing trend after financial crisis and achieving its peak in 2005, amounted to US$8.3 billion.
Moreover, the FDI inflow of Indonesia is instable because of exchange rate volatility. Table 1.4 shows both trend of FDI inflows and exchange rate of Indonesia have the same direction of fluctuation movements. The depreciation of the Raphe happens in Indonesia will lead to an increase in the foreign debt and devalued assets in home currency for those investor companies (Sjoholm, 1999). Hence, these will affect the investor's decision making on investing into Indonesia country.
Table 1.4: GDP, FDI Inflows and Real Exchange Rate of Indonesia in 1990-2010
Gross domestic product (GDP) (US$)
FDI inflows
Real exchange rate, index
Source: World Bank
1.2.3 Thailand
FDI has been an essential component of Thailand's economic development process. Among other developing counties throughout the past four decades, Thailand has becoming the major recipient of FDI inflows (Kohpaiboon, 2003). Even the global experience increase in capital flows, yet there are no clear trends for the region as a whole except for Thailand (Hill & Jongwanich, 2009). The main source that has contributed to the expanding of FDI in Thailand is the increased level of financial development. In Thailand, financial institution or banking industries play an important role in the country's economy. Before the financial crisis in 1997, Thailand economic is considered as a continuous success with an average economic growth rate of nearly 8% per annum from 1960-1996 (Vines & Warr, 2003). This rapid growth of economic is driven largely by the increasing in FDI inflows and exports. Table 1.5 shows the increasing trend of Thailand FDI inflow.
Thailand FDI inflow has grown strongly during the 1990s, and is remained at high levels during 1997-1998 crises. This is because it is able to attract buying opportunities and continued openness. Hence, Thailand FDI inflow has risen sharply amounted from US$2.3 billion in 1996 to US$7.3 billion in 1998. This is due to the "fire-sale FDI," which means crisis cause cheapened asset prices and attractive merger and acquisition opportunities (Hill & Jongwanich, 2009). However, Thailand still faced declining in the early 2000s and it has become strong again until 2007 even there is apparently deterred by political uncertainty. In addition, the global financial crisis in 2008 has affected the FDI inflows causing the FDI inflow to drop dramatically from US$11.3 billion in 2007 to US$4.9 billion in 2009.
Furthermore, Thailand FDI inflow has a fluctuation trend due to exchange rate volatility as explained in Malaysia country. Table 1.5 shows fluctuation in Thailand's both FDI inflows and exchange rate trend. In the event of depreciation of Thai baht, those investment companies with high debt levels will be forced a major restructuring and mainly focus on their home core business (Hill & Jongwanich, 2009). Besides, the currency crisis happens in Thailand in late 1997 has reduced profitability in local market oriented investment (Kohpaiboon, 2003). Hence this will affect the decision of investor to make investment at Thailand country.
Table 1.5: GDP, FDI Inflows and Real Exchange Rate of Thailand in 1990-2010
FDI inflows
Gross domestic product (GDP) (US$)
Real exchange rate, index
Source: World Bank
1.3 Problem Statement
The trend in globalization has led FDI to be substantially globalized in past decades. ASEAN countries are experiencing increasing FDI which is contributed by export and economic growth. Karimi, Yusop and Law (2010) state that the level of FDI inflow to ASEAN countries has been increasing since 2002 and it has led an increasing confidence level of investors in investing and doing business in the region in terms of high FDI. Indonesia, Malaysia and Thailand are the top three developing countries in 2010 shown in Table 1.1. This implies that these three countries are highly depending on FDI. The trade and financial liberalization in many developing countries tend to be integrating the international capital market. This has led to explosive growth in foreign investment (Bansal & Pasricha, 2010). Therefore, ASEAN countries' trade and financial liberalization under ASEAN Free Trade Area (AFTA) are able to accelerate economic integration. The magnitude of possible gains from trade liberalization also has led to a high degree of economic interrelated among ASEAN countries (Goto & Hamada, 1994). Hence, any unfavorable events such as recession in any one of the ASEAN countries will adversely affect the rest of the ASEAN countries.
The macroeconomic variables such as exchange rate and GDP have significant impact on the FDI (Mahmood, Ehsanullah & Ahmed, 2011). The exchange rate volatility is one of the main impacts on FDI especially in developing countries that show extreme volatility (Dhakal, Nag, Pradhan & Upadhyaya, 2010). Hence, the fluctuation on the exchange rate will affect the FDI to be more volatile. According to Anwar and Nguyen (2010), one of the contributions for GDP growth in developing countries is FDI. In other words, the real exchange rate fluctuation in developing nations has generated uncertainty on the economy. The high fluctuation in exchange rate will influence foreign investors' investment decision. This means that the exchange rate volatility increases the complicacy for decision making. This is due to the absolute and relative profitability are unpredictable as well as increasing of uncertainty for cost of production for investment in overseas market. Hence, the high degree in uncertainty of exchange rate movements has not only affected the foreign firm's decision as well as their profit.
Apparently there is an expansive literatures indicating that real exchange rate volatility has direct effect on the FDI inflows. It shows that there is a statistically significant relationship between exchange rate volatility and FDI inflows (Benassy-Quere, Fontagne & Lahreche-Revil, 2001). However, in other literatures such as Furceri and Borelli (2008) argue that there is a weak relationship and significant impact of exchange rate volatility on FDI. The relationship between FDI inflow and exchange rate is yet indistinct, and recent evidence shows that the argument of the different result due to the high open economy countries and period specific. Therefore, the estimated effects on exchange rate to the FDI are varying substantially in terms of magnitude and timing.
Stockman (1980) reveals that there is insufficient evidence to eliminate the adverse effects of inflation on direct investment inflows by neutralizing the effects of inflation on the real exchange rate through nominal exchange rate adjustments. Therefore, the effect of exchange rate volatility on the level of FDI is still far from settled. Besides, FDI is expected to continually grow further due to foreign investors continue searching for more profitable abroad production market. These can be seen in the ASEAN countries such as Indonesia, Malaysia, and Thailand experiencing large inflows of FDI in recent years and is also continually attracting FDI inflows. In additional, these countries are also experience the great deal of variability of exchange rate. These countries' exchange rate are depreciating in the 1990s and appreciating in recent years against dollar (Dhakal et al., 2010). Due to the above phenomenon, we intend to further study to provide better understanding on the relationship between the exchange rate, exchange rate volatility and FDI among the ASEAN developing countries which are Indonesia, Malaysia, and Thailand.
1.4 Research Objectives
Our research objectives consist of a general objective and three specific objectives. Furthermore, our general objective is to study the impact of exchange rate and exchange rate volatility on FDI among three ASEAN countries which are Indonesia, Malaysia and Thailand.
In order to achieve our general objective, there are three specific objectives. First is to examine the characteristic movement of FDI. Secondly, to examine whether there is long run or short run relationship among exchange rate, exchange rate volatility and FDI. Thirdly, to examined and showed the direction of causal effect among exchange rate, exchange rate volatility and FDI.
1.5 Significance of the Research
The contribution in this study is to facilitate policymakers in setting new policies for country development. A good policy is important to stimulate the development of the economy projects. This is due to the fact that it can bring technology transmission and create of job opportunity for country. Besides, a country with appropriate policies is able to attract more potential foreign investors and encourage further foreign investment projects. Hence, this paper is to assist policymaker to identify strategies to attract more FDI.
In addition, policymakers should take macroeconomic and exchange rate stability into consideration in order to increase FDI inflow into the country. Thus, the second contribution in this study is to assist policymaker to understand the effect of exchange rate volatility on FDI in order to make an effective management and planning of exchange rate risk for developing countries.
1.6 Layout of the Research
This paper is organized as follow: Chapter 2 discusses the literature review about the relationship among exchange rate, exchange rate volatility, gross domestic product (GDP), openness and FDI; Chapter 3 describes the data and methodology; Chapter 4 discusses the empirical results; Chapter 5 summarizes this study and the policy implication.
CHAPTER 2: LITERATURE REVIEW
2.0 Overview
Exchange rate is one of the explanatory variables widely used in FDI determination models, however its expected effect on FDI is twofold. There are a number of different arguments on the theoretical literature on exchange rates and FDI. The traditional view argues that investment incentives are not affected by exchange rate level (Blonigen, 1997). Despite the assertions of the traditional view, some studies have stressed the possible effects of appreciation and depreciation of the real exchange rates on the location of domestic and international investment flows.
In most of the results from the previous studies, both theoretical and empirical, indicate that there is indeed a correlation between FDI and exchange rates (Froot and Stein, 1991; Blonigen, 1997; Amuedo-Dorantes & Pozo, 2001; Goldberg and Kolstad, 1995)
Here, the review of past literatures and theoretical framework are discussed.
2.1 Review of Literature
2.1.1 Exchange Rate and Exchange Rate Uncertainty
Many researchers who studied the impacts of exchange rate changes on FDI assume imperfect capital market where borrowers face a premium from external borrowing. Blonigen (1997) creates a model with the assumption of imperfect market. Investors may not have equal access to all markets; relative returns available for investors acquiring a particular asset in host country may be influenced by the exchange rate changes. Buch and Kleinert (2008) study OECD countries in 1977-2002 using robust standard error. The consistency of this hypothesis is shown by observing the appreciation of a country's currency increases the acquisition of firm-specific assets abroad. Furthermore, Benassy-Quere et al. (2001) generated a new variable which is the logarithm of the real exchange rate of the host country relative to that of the home country that result in a positive and significant outcome on 17 OECD countries' FDI in 42 developing economies. Alba, Wang and Pack (2010) also have achieved consistent result where the exchange rate level has a positive and high significant influence on the rate of FDI inflows.
In an influential article on the effect of exchange rates that Froot and Stein (1991) wrote, they came to a conclusion that the logarithm of real value of the dollar has a significant negative influence on U.S. FDI inflow. Generally in their empirical results, foreigners have an advantage and control in purchasing productive corporate assets in the event of currency depreciation. This hypothesis is further elaborated by Klein, Peek and Rosengren (2000) where the role of exchange rates and the wealth effect argument are focused. Their model records and identifies the reaction of international capital movements with the changes in wealth that are affected by the fluctuations of exchange rate. The domestic corporation's bidding power to purchase assets in another country is increased by the appreciation of the exchange rate. Given that if the capital market is not perfectly integrated and free arbitrages are also not allowed, this increased wealth and resulting in bargaining power increase the capability of domestic corporations to obtain assets dominated in the depreciated currency (Udomkerdmonkol, Morrissey & Gorg, 2009).
However there are only a few empirical studies have shown that appreciations of a country currency have no influence or effect on inward FDI (Campa, 1993; Goldberg & Kolstad, 1995). Thus, some ambiguity remains where capital market imperfection as well as information asymmetry cannot clarify the absolute fact that different types of FDI across countries respond differently to exchange rate changes (Pain & Welsum, 2003). The study of Amuedo-Dorantes and Pozo (2001) stated that the consensus between relationship between the exchange rate uncertainty and FDI cannot be reached due to some who claimed that there is a direct relationship, some who showed an indirect relationship and some which even stated that there is no relationship between these two variables.
In addition, these studies also claimed that there are several differences in channel that exchange rates can actually affect FDI flows. These channels signify the wealth position hypothesis that relates FDI to the international exchange markets by the relative wealth of the two nations. As example when there is depreciation of the foreign currency increases the relative wealth of the investing country and makes it profitable to invest offshore.
Most empirical evidences have shown that depreciation of a country's currency will encourage FDI inflows because the relative wealth position of foreign investors increases and the relative cost of acquitting capital will fall which allows them to internally finance more of the investment (Froot and Stein, 1991; Amuedo-Dorantes & Pozo, 2001; Sazanami, Yoshimura & Kiyota, 2003; Kiyota & Urata, 2004). Especially in the study of Froot and Stein (1991) have largely subscribe to 'relative wealth' as the relationship between FDI and exchange rate levels in the case of US inward FDI during 1974-1987. This relative wealth position indicates that real depreciations benefit foreign purchasers of local assets and are related with a rise in inward FDI.
While the other channel, the relative labor cost hypothesis, specifies that the exchange rate affects FDI through relative labor costs. This means that a foreign country with a depreciating currency represents a chance for lower labor costs. Cushman (1985) stresses that currency fluctuations have impact on relative production cost and labor costs. He has used the annual, bilateral direct investment flows from United States to Canada, Japan, Germany, France, and United Kingdom in 1963-1978. A real depreciation of a country currency reduces the cost of FDI as it lowers production costs and also labor wage. The other researchers also show this consistent result in studies on exchange rate effect FDI (Dewenter, 1995; Goldberg & Kolstad, 1995). Some investors pay attention to real exchange rate as an indicator of production costs abroad with FDI representing capital-seeking low-cost labor facilities.
Ito (1999) suggests that home currency per US dollar has mainly a negative impact on Japanese FDI into economies in Asia. Pain and Welsum (2003) find that in the six largest economies (United States, United Kingdom, Canada, Germany, France, and Italy), there is negative significant short-run, but insignificant in the long-run effect in the level of real exchange rate and FDI in developed country. Besides, Dhakal et al. (2010) find the real exchange has negative and significant effect for FDI in developing countries due to the transaction and translation costs which discourage FDI.
Ruiz and Pozo (2008) found that discrete variations in the real exchange rate do not affect FDI in Chile, Argentina, Brazil, Mexico, Columbia, Peru and Venezuela (these countries account for over 85% of the FDI inflow into Latin America). Whether real exchange rate depreciates more or less does not seem to encourage or discourage FDI, countries need not to manipulate the exchange rate if their goal is to promote FDI inflows.
Conversely, the level of real exchange rate uncertainty is significantly influencing the level of FDI received. The result indicates that investors can manage discrete variations in relative prices that arise through discrete exchange rate movements whereas investors have less tolerance to exchange rate movement uncertainty.
However, in reality, every investor have different motivation and it is hard to differentiate among them (Chowdhury and Wheeler, 2008). Hence, in the study of Klein and Rosengren (1992), they tried to investigate the difference on these two channels of wealth position and labor costs and they finally drew the conclusion with only the wealth position channel affecting the firm's investment decision.
Besides that, there are many researchers who have studied the effect of real exchange rate uncertainty on FDI. In the study of Benassy-Quere et al. (2001), the uncertainty of exchange rates is an important factor for investment decision making. Uncertainty is essential to investors as they necessarily look into the future just before undertaking any investments. Recent theoretical literature has highlighted the studies of Dixit (1989), which emphasize the role played by uncertainty in influencing investment decision. The irreversible nature of investment and uncertainty about the future costs and benefits of the investment may cause a wait and see behavior among investors in investment decision making. Dixit and Pindyck (1994) find those FDI investors' concerns about uncertainty as they look into the long term timeline before taking any investments. Thus, FDI behavior will be responsive to the degree of investment uncertainty about economic conditions, future prices and rates of return. This is supported by Lahiri and Mesa (2006) where exchange rates generate extra costs, and firms have to adjust outputs appropriately.
While other recent study, Lin, Chen and Rau (2010) examine empirical evidence from a survival analysis on how exchange rate uncertainty influences the timing of FDI based on the entry by Taiwanese firms into China in 1987-2002. They have developed an integrated model of FDI under uncertainty to demonstrate the impact of exchange rate volatility on the FDI activity of a market-seeking firm versus an export-substituting firm. This model shows that when FDI activity of a market-seeking firm tends to be delayed by exchange rate uncertainty, it can actually accelerate the FDI activity of an export-substituting firm if the degree of risk aversion of the firm is high enough. Goldberg and Kolstad (1995) also study the effect and impact of exchange rate volatility on FDI by using quarterly FDI flows activity between the bilateral partner of U.S. which are Japan, United Kingdom and Canada from 1978 to 1991. They find the effect of exchange rate volatility on FDI is positively related. Their results were also consistent with their theoretical forecast which is after the shock is realized, there is no profitability for those risk averse investors adjusting the productive factors in the short-run. Hence, they concluded that both market demand and real exchange rate shocks tend to lead FDI share increase. Besides, they also found out that a weak effect of a country's FDI outflows tend to decrease due to a depreciated in currency. In addition, Amuedo-Dorantes and Pozo (2001) find that exchange-rate uncertainty may increase foreign direct investment as firms try to reduce exposure to demand fluctuations due to changing terms-of-trade.
Meanwhile, Furceri and Borelli (2008) study the role of exchange rate volatility in explaining the evolution of FDI inflows in the 35 EMU neighborhood countries against the Euro and the Dollar in1994-2004. They find that there is no linear relationship between exchange rate volatility and FDI, the effect of exchange rate volatility significantly depends on the level of openness. The effect of exchange rate volatility is negligible for relatively closed economics (with relatively low FDI flows); whereas there is higher exchange rate stability favors FDI inflows for relatively high open economies (with greater potential to attract FDI and more subject to external shocks).
Besides, Sung and Lapan (2000) have developed model to study the impact between exchange rate uncertainty and risk neutral multinational firm's investment behavior based on the long-run production flexibility. Their results show that the multinational firms are able to obtain profit when investing into two different countries with high exchange rate volatility. Dhakal et al. (2010) also agree to this statement. They further explain that it is possible for multinational firm perceive volatility more towards the depreciation of exchange rates in most of East Asian countries. Hence, these results indicate that the higher the exchange rate volatility, the more the attractiveness of considerable FDI inflows.
Also, many empirical works on the effects of exchange rate uncertainty and FDI inflows has focused on the developed economies. However, few studies that focused on developing countries find a negative relationship between uncertainty of exchange rates and FDI inflows (Bennassy-Quere et al, 2001). Ruiz and Pozo (2008) stated that a high degree of exchange rates uncertainty might dampen companies from making the primary investment in developing countries. They further explain a negative relationship between exchange rate and FDI occurs due to the purpose of investment which is either to bring back the production to their home country or to serve other markets.
Brzozowski (2006), Udoh and Egwaikhide (2008) and Udomkerdmonkol et al. (2009) show a consistent result that exchange rate uncertainty and volatility may negatively influence FDI. For instance, Mahmood et al. (2011) also measure exchange rate volatility using the GARCH model, they discover the presence of negative impact of exchange rate volatility on FDI, while positive impact of exchange rate volatility on GDP, growth rate and trade openness.
Furthermore, Blonigen (2005) states that if the purpose of FDI is to diversify the location of production (increase market share) and to have the choice of production flexibility, then a positive relationship between uncertainty of exchange rate and FDI is to be expected. On the other hand, if the purpose of FDI were either to serve other markets or to bring production back to the home country, then a negative relationship between FDI and exchange rate uncertainty will arise.
According to a priori, the future behavior of an economic variable is uncertain as the possibility of future events cannot be determined. Therefore, the future volatility of an economic variable is understood as a stochastic process that grows over time with a random and a deterministic component. The uncertainty of an economic variable is the unpredictable portion of its volatility (Carrut, Dickerson & Henley, 2000). Both conditional and unconditional measures of volatility have been used in the previous literatures in order to proxy exchange rate volatility.
In addition, Crowley and Lee (2003) study the impact of exchange rate volatility on FDI over the period of 1980-1998 on 18 OECD countries using GARCH (1, 1) model to characterize stochastic volatility in the foreign exchange rate. They found that there may be a threshold effect in the sense that the relationship between exchange rate volatility and FDI is weak or absent if movements in the exchange rate are relatively small, but strong if movement in the exchange rate becomes excessively volatile.
A classic measure used to proxy volatility is the rolling variance, which is an unconditional measure. On the other hand, conditional measures such as the ARCH and GARCH processes are common measures of volatility. In contrast to the unconditional variance of a variable, conditional variance utilizes the previous information to measure volatility. Goldberg and Kolstad (1995) suggested that the rolling variance displays the total variability of the series; however, part of that total variability is foreseeable. It is often argued that unconditional measures of volatility should be stronger measures of total volatility because they include both expected and unexpected volatility. However, when studying uncertainty, the conditional should be a preferable measure because it captures the unexpected volatility (Crawford and Kasumovich, 1996). Therefore the ARCH/GARCH models have been used by many studies that focus on volatility since they produce the conditional variances of a variable.
Many researchers have investigated the effect of the volatility of the exchange rate on FDI with conflicting conclusions. The exchange rate volatility has increased the competiveness between the firms in diverse countries (Chowdhury and Wheeler, 2008). This is due to the fact that firms have the opportunity to enjoy the benefit of cost reduction by moving production (Cushman, 1985). However, the firms have to bear the risk from the exchange rates volatility (Chowdhury and Wheeler, 2008). Hence, the exchange rate volatility will cause a positive or negative impact on FDI. Furthermore, the standard error of the exchange rate has been inspected by Goldberg and Kolstad (1994), who find a significant or positive effect on FDI. However when they investigated the first difference of the standard error of exchange rate, the result turn up to have insignificant or negative relationship with FDI. However, Tomlin (2000) found insignificant results on the standard error of the exchange rate. He reported that the ratio of the FDI announcement-year's average exchange rate to the average rate for the previous 3 years has an insignificant influence on FDI.
In recent study of Chowdhury and Wheeler (2008), there is also a consistent result on positive relationship between FDI and exchange rate volatility. An increase in volatility will increase FDI, while a decrease in volatility will decrease FDI. Furthermore, Osinubi & Amaghioyeodiwe (2009) study the direction and the magnitude of actual inward FDI and exchange rate movement and its volatility from 1970 to 2004. The results show that the effect of exchange rate movement on real inward FDI is positive. This suggests that the depreciation of the exchange rate movement leads to increase in real inward FDI.
Similarly, this view is confirmed by Kogut and Chang (1996) and Ameudo-Dorantes and Pozo (2001) who suggested that the exchange rate movement is a vital determinant of FDI. They find that there is a significant negative effect on FDI between host and home countries (the ratio of the home country's currency to the FDI host country's currency). They argued that the effect of exchange rate movement can be improved if the MNEs have R&D capability together with their prior history of investment in the host country. In addition, Xing (2006) and MacDermott (2008) also concluded that their studies investigate the exchange rate movements on Japanese FDI where an appreciation of Yen encouraged FDI outflows from Japan.
Other exchange-rate-related variables have also been extensively analyzed. Harrison and Revenga (1995) include the adjusted dollar index in their study, an indicator of the degree of distortion of the price structure of the tradable goods. They have discovered that this index has a negative effect; however the first difference of the value has an insignificant effect on FDI. Besides that, in the study of Baek and Okawa (2001), it is discovered that the host currency per US dollar is not a significant determinant of FDI. They also find that the variance between the nominal and Purchasing Power Parity (PPP) rates of the home currency per US$ has a significant negative effect. On the other hand, the variance between the nominal and PPP rates of the host currency per US$ has an insignificant impact.
2.1.2 New GDP or GNP Related Variables
Other than traditional GDP and GNP, several new GDP or GNP related variables have been produced since 1990 which are highly expected positively related to FDI. The effect on the components of GDP or GNP has been investigated extensively.
Edwards (1990) and Harrison and Reyenga (1995) stated that the ratio of the host government consumption to GDP has unclear effects on FDI. Whereas Harrison and Revenga (1995) examined that the mining industrial's production relative to GDP is positively correlated with FDI inflows, while the agricultural industry shows negative correlation. Edwards (1990) discovered that the variable has a positive relationship with FDI inflows, in the case of industrial production to GDP ratio. Bevan and Estrin (2004) report that the share of GDP in private sector of host country which has positive and significant effect on FDI in 12 Eastern European transitional countries. They find that FDI is positively related to both source and host country GDP but inversely related to the distance between the countries and to unit labor costs.
Besides, Wei (1997) uses host population as a proxy for host GDP and reports that it has significant and positive effects on FDI. Tuman (2000) use GDP as a proxy for market size and have an insignificant result in explaining FDI in Latin American countries. Conversely, Trevino, Daniels and Arbelaez (2002) discovered that GDP was not only significant but also an important variable in explaining FDI inflows among Latin American countries. The results were also consistent with the recent study of Ruiz and Pozo (2008) which studied on the exchange rate and direct investment of US into Latin American. Their estimation results shows US GDP and FDI is significantly positive related which implies that US income growth will lead to increase in direct investment flows of US into Latin American.
Moreover, Per capita GDP or GNP is also used in the third generation models. Some studies stressed that there is positive and significant effect of per capita GNP in the home country (Blonigen & Davies, 2000). Furthermore, the purchasing power of a country and its market size reflected on the levels of real GDP of the host country. Several studies show that a country's real GDP on FDI has positive and significant relationship (Wang and Wong, 2007; MacDermott, 2008; Jajri, 2009; Lizardo, 2009; Dhakal et al, 2010). Jajri (2009) found that the GDP for parent and host country are positively influence FDI. Asiedu (2002) uses the logarithm of the expression (1/host GDP per capita) as a replacement of return on investment in the host country. He finds that the ratio of net FDI flows to host GDP for 71 developing countries in the world has a significant positive influence on FDI.
A growing market increases the prospects of market potential and a large market size would generate economies of scale (Root & Ahmed, 1979). While Scaperlanda and Mauer (1969) examine the determinants of the US direct investments in the European Economic Community (EEC), they find that the size of EEC the only significant variable after various simulations. They use the EEC's GNP as a proxy for market size. This hypothesis has been confirmed by a number of empirical studies in developing host countries.
Furthermore, the correlation between host and home GDP or GNP has also been explored. Blonigen and Davies (2000) investigated that the sum of the host and home countries' level of real GDP has a significant positive relationship with FDI. They also report that the square of the variance between FDI host and home country's GDP has a negative significant influence on FDI. Moreover, they examine some interaction variables concerning either the sum of or the difference between home and host GDP, resulted with mainly ambiguous effects.
2.1.3 Openness to International Trade
Openness to international trade has drawn extensive attention in FDI literature after the period of 1990 as FDI determination models. Besides, Ruiz and Pozo (2010) also find openness to trade is an important determinant of FDI. Tuman (2000) find the openness to trade will affect FDI due to the fact that labor intensive assemblies shift from parent companies to its foreign subsidiaries by import capital goods and export final goods back to the parent company or to other countries.
Asiedu (2002) stated that the type of investment will cause different effect of openness on FDI. Normally for a market-oriented firm that seeks to "jump tariffs" are likely to set up subsidiaries in the particular country which is restrictive in trade (difficult export their product to that country) in order to serve local markets (Tuman, 2000). Hence, market-oriented type of investment will tend to lead the less openness country to generate a positive impact on FDI, whereas, export-oriented types of investment lead the open to trade negatively effect on FDI (Asiedu, 2002). They further explained that the export-oriented firm are likely to invest in those countries with more open economy due to imperfections arises trade production which associate with higher export transaction costs. On the other hand, Harrison and Revenga (1995) demonstrated that the dummy of restrictions on trade payments with regard to host current transactions generate negative effect to FDI. Besides, Furceri and Borelli (2008) stated the negative effect of trade restrictions and positive effect of open-trade policies on host countries' FDI.
Most analytical results reflect openness' positive effect on FDI. Harrison and Revenga (1995) have shown the index of openness (the summation of exports and imports to GDP) in host economy is significant positive effect on FDI. Furthermore, Blonigen and Davies (2000) also have discovered a positive effect of home country's index of openness. However, Harrison and Revenga (1995) have found that changes in the index of openness are statistically insignificant.
On top of that, Harms and Ursprung (2002) discovered that FDI inflows have a positive influence when a host trade openness indicator, indicating the low trade barriers, such as low tariff rates, low black market exchange premium and fewer restrictions on capital movements. The open to trade will be positively related with FDI inflows due to the fact the transnational corporations who seek for more open and trade economies for resource-seeking operations and allow to import and export easily by integrate their production internationally (Dhakal et al. 2010). They further explain that the openness to trade is a vital determinant of FDI in the case of East Asian countries because these countries have been relatively open to investment and trade historically. However, Lee and Masfield (1996) applying openness indicators of the host country practices (price controls, free import, profit repatriation controls) into their model, they get insignificant results. Also, Bandelj (2002) concludes that there is an insignificant effect on FDI inflows to the openness of host government's FDI policy in the Central and Eastern Europe.
Besides the three groups of the most commonly used explanatory variables cited above, other several factors are studied in FDI literature: 1) science and technology, 2) industrial characteristics (company characters, profits, industrial shipments, and industrial organization), 3) international trade variables (labor emigration, net energy imports, trade open year, terms of trade, export comparative advantage indicator, dependence on imported raw material), 4) new financial variables (money supply, stock market, credit and debt, and capital), 5) human capital (population and density) and employment, 6) new public finance variables (government expenditure, capital account), 7) social-political variables (transition, diplomatic variables, and culture), 8) geographic locations and distance, 9) natural resources and environmental protection, 10) various dummies (time and country). All these new studies are the contributions of various researchers have widened our understanding the scope of FDI.
CHAPTER 3: METHODOLOGY
3.0 Overview
This chapter explains the theoretical model and methods that have been used to estimate the impact of exchange rate volatility in FDI. In our study, the selected developing ASEAN countries are Malaysia, Indonesia and Thailand because these countries have the highest FDI in 2010. This chapter consists of three important sections. Section 3.1 describes the data. Section 3.2 describes the variables used in this study. Section 3.3 explains the empirical model test to conduct our analysis on FDI. Section 3.4 draws a summary on all the mentioned subtitles and provides a linkage to the next chapter.
3.1 Data
In this study, we have obtained quarterly data for each of the country in order to estimate the impact of exchange rate and volatility exchange rate on FDI. The observation period that is used in this study is from 1981 until 2010, where our sample size is n =124. The data of Consumer Price Index (CPI), export and import for Malaysia, Indonesia and Thailand are obtained from IMF database. The data for exchange rate, FDI and GDP in each of the countries are obtained from the Oxford sources. All the data in this study will be expressed in million units of the US dollar. CPI and GDP values are in the base year 2005.
3.2 Variables
The dependent variable is FDI inflow in country i in time t. There are four independent variables are used in this study, namely Gross Domestic Product (GDP), Real Exchange Rate (RER) denominated U.S. dollar of country i in time t, Exchange Rate Volatility (RERV) in country i in time t and Openness (OPENESS) in country i in time t. Real FDI (presented in percentage form) is computed through the nominal FDI divide by GDP deflator. Whilst, RER is computed through the exchange rate in term of USD relative to home country and multiply by the total value of the CPI of home country divided by the CPI in US. Moreover, OPENESS is derived from the sum of nominal import and nominal export relative to the nominal GDP. The RERV will be explained further in the next section to measuring the volatility of the exchange rate.
3.3 Research Methodology
3.3.1 Measuring the volatility of exchange rate
Exchange rate volatility is one of the important variables in explaining the FDI in Malaysia, Indonesia and Thailand. To examine the impact of the RERV on FDI, RERV needs to be derived from the conditional variance of the RER through generalized autoregressive conditional heteroskedastic (GARCH) model. GARCH and ARCH model are developed by Bollerslev (1986) and Engle (1982). The purpose of this model is to measure the volatility of the RER. RERV is derived for each of the selected country which is Malaysia, Indonesia and Thailand. In our study, GARCH (1, 1) model is constant in each country for the convenience in comparing among the RERV in the selected three countries. The term of (1, 1) in GARCH (1, 1) refers to the first-order autoregressive GARCH term and subsequence by the first-order moving average Autoregressive Conditional Heteroskedastic (ARCH) term. The GARCH can predict recent period variance through forecasted variance from the last period while the ARCH is the volatility information that is obtained from previous period.
We select Autoregressive Moving Average Model, ARMA (p, q) model as the dynamic model in our study. ARMA is the combination of the Autoregressive Model, AR (p) model and Moving Average Order, MA (q) model. P in the ARMA model can be defined as the number of autoregressive order while q is the number of the moving average order. Both p and q is involved in the autocorrelation function (ACF) and partial autocorrelation function (PACF) correlogram. For illustration, ARMA (1, 1) is applying in our study and it defines linear difference equations with constant coefficients. ARMA (1, 1) can be determined through the ACF and PACF correlogram.
The ARMA (1, 1) process for RER is express as follow:
Yt = θ + α1Yt−1 + β0Ut + βUt−1 ----------------------------------------------(1)
Yt = Real Exchange Rate
θ= constant term
There are some elements that need to be fulfilled such as the dependent variable (FDI) must have stationary process and the error term is assumed having white noise process. In other words, the error term follows the identical independent distributed, stationary process, constant with the previous error term and zero mean and variance. In order to examine the characteristic of stationary movement in residual, unit root test is applied. The purpose of unit root test is to determine the characteristic of residual which is constant and stationary and can fulfill the assumption of white noise process. In other words, stationary of residual is important in explaining the estimate model and to make sure the estimated model is unbiased.
3.3.2 Concept of Unit root test
According to Elliot, Rothenberg and Stock (1996), unit root test is performed to determine the data stationary characteristic. Computing a unit root test is essential to find out whether the variables in time series are stationary or non stationary. This is because the stationary or non-stationary characteristic of a series will have a strong influence on its behavior and properties.
In the time series literature, stationary can be defined as the mean, variance and covariance of series that are constant over the period. On the other hand, non stationary has a time varying mean or variance in time series. Once the series are non stationary, the estimates from non stationary variable may generate a spurious regression. The traditional t-ratios will not follow the t-distribution; hence the hypothesis test about the regression parameters will no longer be valid. In an effort to determine the data stationary characteristic, we will perform Augmented Dickey-Fuller (ADF) test.
3.3.2.1 Augmented Dickey Fuller unit root test
Augmented Dickey Fuller unit root test is developed by Dickey and Fuller 1984, which is the modified the version of Dickey-fuller (DF) test. This ADF test is conducted by augmenting the equations by adding the lagged length of the dependent variable Yt into the model (Sephton, 2008).
In our model, ADF test is use to estimating the following regression:
Yt = µ + βt + γYt-1 + + εt --------------------------------- (2)
Yt = FDI, GDP, RER, RERV, OPENESS
µ= Constant
βt = The coefficient of time period
k = the number of lag under of autoregressive process
εt = error term
t = time or trend variable
In order to compute ADF test, we determine the Schwarz Info Criterion (SIC) in assisting us to obtain the optimal lag length. Moreover, to estimate whether stationary or non stationary in ADF model, hypothesis in ADF unit root test is H0: β1 is non stationary while H1: β1 is stationary. The null hypothesis will be rejected if the test statistic is less than the critical value, otherwise do not reject null hypothesis. The critical value can be obtained from the t statistical table. Hence, if we then reject the null hypothesis, we are confident that the series has a stationary.
3.3.3 Vector Autoregressive models
Once the unit root test has showed that our variables is stationary in a way that the FDIt is stationary at level I (0), Xt variables is stationary at first difference I (1). FDIt is stationary at first difference I (1), Xt variables is stationary at level I (0). FDI t and Xt variables are stationary at level I (0) therefore we can continue to estimate the VAR model to be more precise. In other words, VAR model is free from unit root in above circumstances.
Vector autoregression model (VAR) is a statistical model that is used to capture the linear independent variables among time series. The VAR model is essential especially for describing the dynamic behavior of economic and financial time series and even useful for forecasting (Agrawal &Shukla, 2011). Assumption of VAR consist of expected residuals are zero, and the error terms are not autocorrelated. In other words, if the error terms between the variables are uncorrelated which mean the estimation will be unbiased and efficient.
In our study, we use VAR models to examine the impact of exchange rate volatility of each country on FDI in three countries such as Malaysia, Indonesia and Thailand. The term autoregressive is due to the appearance of the lagged value of the FDI on the right-hand side.
Given the equation below is to estimate by OLS regression: ï€
FDIt = α ++++
++ ε1t ------------------------------------------------ (3)
GDPt = λ ++++
++ ε2t ------------------------------------------------ (4)
RERt = αʹ ++++
++ ε3t ------------------------------------------------ (5)
RERVt = λʹ +++
+ ++ ε4t ---------------------- (6)
OPENESSt = η +++
+ ++ ε5t --------------------- ---- (7)
α, β and others that represent this two are known as parameter. ε are the stochastic error terms, t is time or trend variable, K is the optimal lag variables. Determined the optimal lag length is an essential process because as what we know includes too many lagged length in model will reduce the degrees of freedom, while too few lags in our model will lead to specification errors. Akaike (1978) has shown the SIC is outperform than the AIC in estimating the order of an autoregressive model. Once the VAR model is estimated, therefore we can persist to other steps that will be discuss later.
3.3.4 Granger Causality Test
After construct the VAR model with all variables and residual value in stationary form, we can proceed to the Granger Causality test. The Granger Causality test had proposed at the year of 1969 by Granger. The purpose of the Granger Causality test in our study is to examine whether the FDI is causing RERV or is RERV that causes the FDI. Other than that, it also determines the causality cause between the FDI and RER. Either FDI is causing RER or is RERV causing FDI. Moreover, Granger Causality Test is taking into account of the conditional mean. Based on the VAR estimation, the null hypothesis in this case is FDI does not cause RERV while the alternative hypothesis is FDI is causing RERV. Null hypothesis is rejected when Wald F test is greater than critical value. When null hypothesis is rejected we can conclude that FDI cause RERV in short run in the VAR estimation model.
When we have applied the Granger Causality Test in our study, we should understand the concept and some assumptions of Granger Causality Test. There are two concepts in Granger Causality Test. First, the future cannot affect the past and the past affects the present or future. Hence, past comes before the future. Granger Causality test is only to determine whether one incident happens before another incident and help to predict it. It does not imply the effect or the result. Second, the past can help in predicting the future because of the available of past historical information of the other variable in the time series model.
There are some assumptions need to be fulfilling before apply the Granger Causality Test. First, the variables must be stationary or there is no pattern existing in the time series movement. However, if the variable showing the non-stationary, the first difference of the variables is taking place to transform the variable to stationary. Second, the direction of causality may strongly depend on the number of lagged terms included. For example, Schwarz information criterion (SIC) can be computed to determine the number of lagged term in the case of distributed lag models. Third, the error term is assumed uncorrected when involved in the causality test. Last, compute the Wald F test in order to the testing the causality causes between FDI and RERV and FDI and RER. The Wald F test formula is express as:
3.3.5 Variance Decomposition
After determine causality cause among the variables with Granger Causality Test, we further investigate further detail on the impact on system variables of typical shocks through Variance decomposition in our VAR estimated model. Besides, impulse response function (IRF) and variance decomposition (VDC) are the two short run analysis in time series analysis (Adam & Tweneboah, 2008). According to Sims (1986), forecasting variance has existed when residual value is related in the variable optimally forecast from its own lagged value. Thus, VDC is usually applied to explain most of forecast error variance in the short run timeframe.
VDC or forecast error variance decomposition is used to aid in the interpretation of a VAR model. It reveals its shocks to one variable that have significant impacts on another variable or how important variation of FDI explaining in variation on endogenous variable such as GDP, RER, RERV and OPENESS. VDC indicates the amount of information each variable contributes to the other variables in the autoregression. VDC determines how much of the forecast error variance of each of the variables can be explained by exogenous shocks to the other variables. It enables us to determine the proportion of one of the variable relation for by its own shocks and the shocks to the other variables.
Moreover, VDC can determine the direct or indirect impact between the endogenous and exogenous variables. However, there is one drawback in VDC which is it is unable to account the direction or the trend of impact in a time series model. In our study, we focus on the forecast error variance in FDI explained by shocks to GDP, RER, RER, and OPENESS.
3.3.6 Impulse Response Function (IRF)
Based on the VAR equation 3,4,5,6 and 7 that mentioned above, we had adopted the Impulse Response Function to study the responses of the effect of a volatility exchange rate and exchange rate on FDI in future or current period of time. In short, impulse responses functions refer to the impact of a dynamic system are presumed to be responses to the external change (Kassim, Majid & Yusof, 2009)
As we know VDCs also can also show the impact of shocks to one variable on another variable. However, they do not show the direction of these impacts. Therefore we are applying IRF in our model to detect the pattern and direction as positive or negative sign of the transmission of the shock to the variables. For example, there is a significant positive relationship between the RERV to FDI. This indicated that the increase of fluctuation or shock in the volatility of exchange rate will expect to higher positive impacts on foreign direct investment in certain country at a given period.
We can identify the trends or direction of the impulse response on the variables based on the graph. When there are high positive effects on the variables, the impulse response to other variables will tend to fluctuated. In addition, mean squared deviation can be used to estimates the accuracy of impulse response function. Therefore we able to know whether there is a strong impulse response between these variables.
3.4 Conclusion
This chapter provides the outline of data collection method, variables, and also concept of research methodology that employed in our study. Constructions of this chapter is able to provide a better understanding to readers on how our study is constructed in terms of data, empirical model, measuring volatility of exchange rate and causality between the variables. Data that we used in this study is secondary data from International Monetary Fund (IMF) and Oxford sources. The sample period in our study is from 1981 to 2010. Next, the description on statistical methods has been explained including ARCH and GARCH to measure the volatility of exchange rate on Foreign Direct Investment, unit root test, estimated VAR model, Granger causality test, Variance decomposition and Impulse response function. In our study, we suggest this chapter can provide useful information to university students, researcher, economist and government. Next, the empirical results for the test will be further explained and interpret in the following chapter.
CHAPTER 4: RESULT AND INTERPRETATIONS
4.0 Overview
This chapter shows the result and interpretations of the relationship among the exchange rate, exchange rate volatility and FDI for the Indonesia, Malaysia and Thailand for the period 1981-2010. The first section of this chapter is the results of descriptive statistic for all the series. Next, the second section explained the result of unit root test with ADF test. Thirdly, estimation in VAR and explanation of the results in Causality analysis, IRF analysis and VDCs analysis.
4.1 Descriptive Statistic
Table 4.1 is the result of descriptive statistic for all the series in Indonesia. The table revealed that maximum and minimum values of R
