CHAPTER- 6 DETERMINANTS OF FOREIGN DIRECT INVESTMENT IN INDIA AND CHINA FDI usually represents a long term commitment to host country and contribute significantly to gross fixed capital formation in developing countries. FDI has several advantages over other types of capital flows, in particular its greater stability and the fact that it would not create obligations for the host country as has been observed in the context of the Asian financial crisis of 1997-98 (Cho, 2003). The ongoing process of integration of the world economy has led to a significant change in the attitudes of the host countries with respect to inward foreign direct investment (FDI). FDI is no longer regarded with suspicion by the developing countries and controls and restrictions over the entry and operations of foreign firms are now being replaced by selective policies aimed at FDI inflows, like incentives, both fiscal and in kind (Banga, 2003). Emerging issues in the areas of foreign direct investment are an essential part of the core process of globalization. FDI can play a key role in improving the capacity of the host country to respond to the opportunities offered by global economic integration, a goal increasingly recognized as one of the key aims of any development strategy. Virtually all countries are actively seeking to attract FDI, because of the expected favourable effect on income generation from capital inflows, advanced technology, management skills and market know-how (Cho, 2003). The determinants of the FDI are numerous. Whether particular action of investor or government is responsible for increase or decrease in the investment for a given period is treated as determinant. There is not a single variable which would influence investment to rise or fall but it is comprised of a set of variables. It would be very valuable to review the key determinants and factors of FDI based on the theories of international investment. The FDI theories are categorized into two parts in order to know the theoretical determinants of FDI. (a) Theories based on Perfect and (b) Theories based 82
on Imperfect market. The perfect market assumes that there exists competition for investment, equal opportunity, and there is equal return on investment across the countries. Perfect competition within the industries implies that there are numerous firms manufacturing same items of same quality and all industries have equal rate of return and tax rate. According to imperfect theory, the financial markets are never perfect. The information needed to take rational decision is rarely available. The risk associated with different level of investment also differs. The investment schedule of the investing firm depends upon rate of return in imperfect market. The industrial organizations across the world are neither identical nor face same problems at a same time. A) Theories based on Perfect Markets: Differential Rate of Return Portfolio Diversification Market Size Resource Location B) Theories Based on Imperfect Market: Industrial Organization Internationalization Liquidity Foreign Exchange Rate Political Stability Tax Policies Government Regulations Trade Policy (Gedam, 1996) 83
Table 6.1: Hypotheses of FDI Theories and Their Evidence Theory Concerning Hypothesis of the Theory Evidence Existing Studies Differential Rate of FDI flows from low to high rate return region Evidence don t support the Gedam (1996); Azam & Return hypothesis Lukman, (2010) Portfolio Diversification Expected risk and return determines the flow of FDI Weak support to validity of theory Agarwal, (2001) Market Size (GDP) GDP growth is proxy for potential market size for sales, which in turn determines flow of FDI Evidence support the theory Coughlin and Segav, (2002); Azam & Lukman, (2010) Resource Location FDI flows will be adversely affected if the natural resources are highly protected. Evidence support the theory Zhang, (2001); UNCTAD, (1998) Industrial Organization Structural imperfection determines the FDI flows Evidence support the theory Gedam, (1996) Internationalization FDI is result of firms replacing transaction cost with Evidence don t support the theory Zhang (2001) internationalization Liquidity Relationship between internal cash flows and Evidence support the theory Chopra, (2003) reinvestment determines FDI flow, i.e. cost of internal funds is lower than cost of external borrowings makes FDI to grow. Foreign Exchange Rate Relative strength of currency determines the FDI flow Evidence support the theory Shan, (2002); Dees, (1998); Cheng & Ma (2008) Political Stability Political, economic and social stability makes FDI to occur and instability deter FDI Mixed evidence or no evidence to support the hypothesis Asiedu, (2002); Ali & Guo, (2005); Zhang, (2000) Tax Policies Tax affects net return on investment therefore tax system Weak support to validity of theory Zhang, (2000) determines FDI. Government Regulations Favourable regulations make the FDI to occur Evidence support the hypothesis Chopra, (2003) Openness More open economy to outside external trade can attract more FDI. Evidence support the hypothesis Azam & Lukman, (2010); Chopra, (2003) The Level of External Indebtedness More burden of repayment and debt servicing making the country less attractive for foreign investor Evidence support the hypothesis Botric and Skuflic, (2005), Azam & Lukman, (2010); Foreign Exchange Reserves Chopra, (2003) More Reserves has positive impact on FDI Evidence support the hypothesis Cheng & Ma, (2008) 84
6.1 Expected Theoretical Relationship between FDI and its Determinants There are so many determinants of FDI in the economy as suggested by existing literature available on this issue. There is need to know the expected relation between FDI and these determinants before doing empirical investigation regarding relationship of FDI and some variables taken in this study so as to find main determinants of FDI in India. (i) (ii) Market Size: Market size which is measured in terms of GDP is expected to have positive relationship with FDI. Countries having more GDP growth rate can attract more FDI inflows. Market oriented FDI aims to set up enterprises to supply goods and services to the local market. This kind of FDI may be undertaken to exploit new markets. The market size of host countries is very important location factor for market oriented FDI. The general implication is that host countries with larger market size, faster economic growth and higher degree of economic development will provide more and better opportunities for these industries to exploit their ownership advantages and therefore, will attract more market-oriented FDI. Even for export-oriented FDI, the market size of host countries is an important factor because larger economies can provide larger economies of scale and spill-over effects (OECD, 2000). Portfolio Diversification: The diversification of portfolio is also considered to be another determinant. The approximate mix of bonds, securities, stock, debenture, depository receipts, etc. refers to portfolio investment. The maturity of these instruments may vary from few months to few years. The concern of an investor is for these instruments at a time of risk perceptions. It implies that the investors are able to invest in or take out their capital for diversification of their portfolio assets due to perceived risk in a country. The higher is the perceived country risk due to political, economic and financial changes in one country, an investor would like to take out his capital out of the country (Gedam, 1996). 85
(iii) Resource Location: Location- specific determinants have a crucial influence on a host country s inflow of FDI. The relative importance of different location-specific determinants depends on at least three aspects of investment: (1) The motive for investment (e.g., resources, market or efficiency-seeking), (2) The type of investment (e.g., services or manufacturing), and (3) The size of the investors (small and medium MNEs or large MNEs) Natural resources protected from international competition by imposing high tariffs or quotas, still play an important role in attracting FDI by a number of developing and developed countries. The theoretical analysis concludes that policy related variables and economic determinants together explain the variations in the FDI inflows in country. Empirical analysis concludes that the variables considered for the study are more significant in China as compared to India. In India, Long term debt is an important factor in attracting FDI but in China Foreign exchange reserves and Sum of exports and imports (Exim) have more influence on FDI. These flows will be adversely affected if the natural resources are highly protected (UNCTAD 1998). (iv) (v) (vi) Differential Rate of Return: This theory explains mostly the held belief that the FDI flows to that country which has relatively higher return on the investment. No investor would like to invest if the rate of return on investment is low. Therefore, the flow of capital will be in those countries which ensure the highest possible rate of return (Gedam, 1996). Foreign Exchange Reserves: The high level of foreign exchange reserves in terms of import cover reflects the strength of external payments position and help to improve the confidence of the prospective investors. Therefore, a positive relationship is postulated between the foreign exchange reserves and the inflow of foreign direct investment (Chopra, 2003). Internationalization: Internationalization refers to minimize or eliminate cost of external transaction by increasing transaction within subsidiaries. This theory explains that FDI is an outcome of need to lower the cost of transaction. In other words, need for internationalization of transaction cost determines the FDI inflows. The internationalization of transaction cost is achieved through FDI investment in subsidiary to eliminate high cost of 86
transaction or replace high cost transaction through low cost when it is impossible to eliminate (Gedam, 1996). (vii) (viii) (ix) (x) (xi) Openness: Openness of a country is generally measured as the proportion of exports and imports to the GDP (Trade/GDP). The more an emerging market tries to open its economy to outside external trade, the more this host country can attract FDI. Export oriented FDI depends upon liberal trade policies reflected in openness of the country as the TNC is not interested in market seeking behaviour initially and openness helps it in importing components, capital goods, and raw material (Zhang, 2001) Government Regulations: This consists of rules and regulations governing the entry and operations of foreign investors. FDI cannot take place unless it is allowed to enter in a country. Its potential relevance is evident when policy changes sharply in the direction of more or less openness. It should be noted, however that policy changes in the direction of openness differ in an important way from those in the direction of restriction. Open policies are basically intended to induce FDI while restrictive policies such as sweeping nationalization of foreign affiliates, can effectively close the door to FDI (Chopra, 2003) Political Stability: The reliability and political stability determines the FDI inflows. TNCs prefer stable government so that their investment is protected. Political instability may be in the form of negative attitude of the government toward TNCs, non allowance of fund transfer, currency convertibility, war, bureaucracy and corruption. Political stability can also be measured by number of changes of democratically elected governments. Asiedu (2002) does not find any evidence relationship between FDI and political stability (Gedam, 1996). Tax Policies: Fiscal policies determine general tax levels, including corporate and personnel tax rates and thereby influence inward FDI. Other things being equal a country with lower tax rates should stand a greater chance of attracting FDI project than a country with higher rates. It is difficult to ascertain how much influence it can have on the total inflows of FDI. (Chopra, 2003). Inflation: Low inflation rate is considered to be a sign of internal economic stability in the host country. High inflation rate indicates incapability of the 87
government to balance its budget and failure of the central bank to conduct appropriate monetary policy. Changes in inflation rates of the domestic or foreign country are anticipated to alter the net returns and optimal investment decisions of the MNEs. It is expected to give negative impact on FDI (Banga, 2003). (xii) (xiii) (xiv) Industrial Organization: Industrial organization theory states that firm specific advantages, competition capabilities, managerial skills and practice etc. are some of the crucial points for industrial organization to survive. The relative advantages to TNCs in terms of these points make FDI to flow to a country of their choice (Gedam, 1996). The Level of External Indebtedness: The level of external indebtedness means the net external assistance to India in the form of loans. It is expected to have a negative impact on FDI inflows. The level of indebtedness shows the burden of repayment and debt servicing on the economy, thus making the country less attractive for foreign investors (Chopra, 2003). Foreign Exchange Rate: It is the rate at which one currency may be converted into another. In other words it is the relative strength of the domestic country in relation to the foreign country. High volatility of the exchange rate of the currency in the host country discourages investment by the foreign firms as it increases uncertainty regarding the future economic and business prospects of the host country (Banga, 2003). FDI arises mainly from the activities of TNCs that operate across the countries. The literature on FDI determinants indicates that TNCs would allocate their investments among countries in order to maximize their profits at low level of risk. However the profit earned by TNCs depends on three factors: Factors within the firm that enable it to grow and expand more successfully. Factors in the host country that make the country as the best location to produce across countries. Factors associated with the firm s trade-off between FDI and exports (Gregorio and Lee, 2002). 88
The relative importance of FDI has increased more since mid 1980s and it is the largest single component of private capital flows to developing countries. The global competition for FDI among developing economies is increasing. The two large emerging economies in Asia wherein this competition is evident are India and China. Both these economies are now getting increasingly integrated with the global economy as they open up their markets to international trade and investment inflows. China has been globalizing at a particularly rapid pace, accompanied by a many fold increase in net inflows of FDI over 1980-2000 period. India began to liberalize its economy about a decade later than China. However, India s market-oriented economic reforms undertaken in 1991 which were directed towards increased liberalization, privatization and deregulation of the industrial sector, and to re-orient the economy towards global competition by reducing trade barriers, and gradually opening up its capital account, has led India to increasingly become a favourable destination for foreign investors (Srivastava and Sen, 2004). Despite good prospects for foreign direct investment, FDI inflows were limited in India as compared to China. It is necessary to evaluate the policy instruments that should be adopted by India to attract FDI and to recognize the locational factors through which the country can influence the flow of FDI. So, some major factors particularly affecting FDI inflows in India have been assessed in order to estimate the determinants of FDI for India. Zhang (2001) estimated the effect of location characteristics and government policy on FDI. It was found that China s huge market size, liberalized FDI regime and improving infrastructure are attractive to multinationals. Dunning (1981) determined the effect of three factors such as Ownership, Location and Internationalization on FDI. Cheng and Kwan ( 2000) estimated the effects of the determinants of foreign direct investment (FDI) in 29 Chinese regions from 1985 to 1995, and found that large regional market, good infrastructure, and preferential policy had a positive effect but wage cost had a negative effect on FDI. Taylor (2000) reports that the incentives by the government of the country have positive affect on attracting FDI. 89
Wilhelms (2004) explore d the determinants of FDI in 67 emerging economies and found that the factors like government fitness, market fitness, educational fitness, and socio-cultural fitness are important determinants of FDI. Brewer (1993) discussed the direct and indirect effects of government policies on FDI and concluded that same government policy may have both positive and negative effects on FDI, therefore empirical evidence on the impact of selective government policies on FDI inflows is ambiguous. Grubert and Mutti (199 1) and Loree and Guisinger (1995) found that investment incentives positively effect inward FDI flows and performance requirements imposed by the host governments gives negative impact on FDI inflows. Devereux and Griffith ( 1998) revealed that fiscal incentives plays primary role in attracting export oriented FDI, while role played by other incentives was found to be secondary. Nunnenkamp (2002) argued that traditional market related determinants are still dominant factors attracting FDI in spite of observing so many changes since 1980s. Gastanaga et. al (1998), Chakrabarti (2001) and Asiedu (2002) have explored that the variables like openness and regional agreements in trade are very significant factors in attracting FDI inflows. Blomstrom and Kokko (1997) analysed the direct and indirect effects of regional trade agreements (RTA) on FDI inflows and concluded that lowering interregional tariffs can lead to expanded markets which results in increase in FDI inflows. 6.2 Choice of Variables From the through review of existing literature and some of the studies discussed above, it has been observed that a lot of work has been accomplished on finding the effect of reforms, change in economic policies, socio-political environment and cost of capital on FDI inflows in India. There is a strong consensus in the literature that multinational corporations invest in specific locations mainly because of strong economic fundamentals in the host countries like availability of infrastructure, stable macro economic environment and favourable policies. 90
The studies conducted on empirical investigation of the some important factors like foreign exchange reserve (RES), Openness (OP i.e sum of Exports and Imports as a percentage of GDP), Long Term Debt (LTD) are really inadequate. The variable openness is suppose to give positive influence on FDI inflows as every country adopting more liberalised policy regime is able to attract more FDI inflows but Tolentino (2009) found no relationship between FDI and Openness in China. So, it becomes imperative to study this variable on account of this conflicting view. The emerging economic giants, the BRIC (Brazil, Russian Federation, India, and China) countries hold the largest foreign exchange reserves globally. India and China are amongst the top 10 nations in the world in terms of foreign exchange reserves. These countries have considerable position among the ten largest gold holding countries in the World (Economic Survey 2009-10). The growing importance of this variable in these countries requires to estimate the influence of foreign exchange reserve on FDI. The variable Long Term Debt shows the dependence of the country on external sources which may cause doubt regarding the financial credibility of the country that influences negatively the inflows of FDI in that country. But now a day s developing country may use this external debt for the development of infrastructure and mobilisation and efficient use of physical and financial resources of country which in turn may attract foreign investors. So, it is important to study the effect of the variable to have more clarity regarding this variable in India and China. Very few studies have evaluated the impact of inflation but there is serious need to evaluate extensively the impact of this ever changing essential variable with the updated time period in India. Pan, (2003) has found positive impact of exchange rate in case of China while negative impact of this variable has been reported by Ali and Guo (2005) in China. Moreover Tolentino (2007) has investigated no relati onship between FDI and exchange rate in china. It requires the verification of its true impact on FDI inflows in India and China due to lack of general consensus. Market size is a very important factor required to be checked in the current time period. So, it has been reflected through selection of the variable GDP as a determinant of FDI inflows in India. All the facts and logics given above reveal that FDI is not only influenced by the regulatory framework but also by many other economic factors. Keeping in view 91
the findings of existing literature, an attempt has been made in this chapter to trace the effect of some selected important economic variables on FDI inflows in India and China. 6.3 Variables, Data Source and Period of the Study This chapter explores the determinants of foreign direct investment which influence FDI inflows in India and China. Quarterly data for the period 1990-91 to 2008-09 has been used in India while annual data covering the period from 1976-2008 has been used in case of China. The data has been taken from the Handbook of Statistics of Indian Economy published by Reserve Bank of India and from World Development Indicators and World Development Reports published by the World Bank for India and China respectively. The variables examined for the study are Gross domestic product (GDP), Foreign exchange reserves (RES), Openness (OP i.e sum of Exports and Imports as a percentage of GDP), Long Term Debt (LTD), Exchange rate (EXCH) and Inflation (INF). Description of Variables: Variables LNFDI LNEXCH LNGDP LNINF LNLTD LNOP LNRES Description Natural Log of Foreign Direct Investment Natural Log of Exchange rate Natural Log of Gross domestic product Natural Log of Inflation Natural Log of External indebtedness Natural Log of Openness (sum of Exports and Imports as a percentage of GDP) Natural Log of Foreign exchange reserves 92
6.4 Statistical Diagnostic (India) As we know that economic time series move together therefore if we include all the explanatory variable in the regression equation there may be the problem of multicollinearity. Before proceeding to further analysis, the existence of multicollinearity among the independent variable had also been examined. For this purpose, Pearson s correlation matrix has been formed that signalled high correlation among various independent variables i.e. causing the problem of multicollinearity. Using Ordinary Least Square linear equation, the explanatory variables are regressed to test the significance of these variables. The multiple regression analysis has been used and the regression results have been reported in Table 6.2. In the analysis a combinations of variables like GDP INF OP EXCH have been found to be statistically significant in India, while coefficients of LTD and RES do not have significant t-value. The value of F is found to be significant in all the equations which show the significance of the model. The value of adjusted R 2 is found to be 0.89 which indicates the percentage variation in FDI due to the combination of variables taken in the study. Table 6.2 Regression Results (FDI as Dependent Variable) (India) Variables Coefficients Std. Error t-statistics Prob. VIF C -4.164178 5.410556-0.769640 0.4441 - GDP -1.650801 0.721537-2.287893 0.0252 17.801 INF 8.263051 1.408933 5.864760 0.0000 57.807 LTD 0.341063 0.209984 1.624231 0.1089 4.700 OP -1.165718 0.567235-2.055089 0.0437 8.786 RES -0.178360 0.346542-0.514687 0.6084 35.484 EXCH -1.578592 0.835465-1.889477 0.0630 6.150 Adjusted R 2 0.89 F-statistics 109.61 Prob. (F-statistic) 0.000 Durbin-Watson 1.000 93
Further the analysis also reveals that the value of Durbin-Watson statistics is very low which reveals the presence of autocorrelation and the value of VIF for all variables is very high which points out the interrelationship between these explanatory variables indicating the existence of multicollinearity. But Adjusted R 2 is very high indicating the spurious regression due to above mentioned problems. Therefore to overcome these problems, Cointegration technique has been applied to find out the factors influencing FDI, 6.5 Econometric Methodology Before estimating any relationships between FDI and its explanatory variables, there is a need to check the stationarity of each series. As non stationary time series will necessarily contain permanent component. Therefore mean and variance of this time series will depend on time. So, it may give spurious regression results. As per description given above about the variables in study, these have been taken in logarithmic form to make them stationary at lesser order of integration. Further the coefficients of log linear model provide elasticities which can be interpreted in the form of percentages and thus free from quantification of variables under evaluation. The stationarity of these seven variables LNFDI, LNGDP, LNEXCH, LNINF, LNOP, LNLTD, and LNRES has been tested by applying formal Unit root procedure i.e. the Augmented Dickey-Fuller (ADF) te st. Cointegration analysis has been applied to study the long run relationship among these variables for estimating impact of these variables on FDI in India. 6.6 Results and Discussion The ADF test results for the seven variables involved in the analysis have been presented in Table 6.2 and in equation form for lucid explanation. It has been observed that the null hypothesis of presence of unit root has been rejected for all the first difference variables specified. This shows that all variables exhibit integrated order one. This means that the series are non-stationary in level but stationary in firstdifferences. 94
Table 6.3: Augmented Dickey Fuller Test Results Unit Root Tests at Logarithmic levels (India) Sr. No. Variables Without Drift and Time Trend With Drift With Drift and Time Trend 1 LNEXCH 1.6742-1.1933-1.6029 2 LNFDI 1.4655-1.5516-2.4106 3 LNGDP 4.2041 0.4483-2.4804 4 LNINF 2.0729-2.3701-3.0247 5 LNLTD 1.3970-1.0846-3.4890 6 LNOP 2.4279-0.2746-1.9671 7 LNRES 3.9227-1.7300-2.9310 Unit Root Tests at First Differences Sr. No. Variables Without Drift and Time Trend With Drift With Drift and Time Trend 1 LNEXCH -7.3690** -7.5214** -7.4697** 2 LNFDI -9.6118** -9.9363** -9.9070** 3 LNGDP -1.9955* -5.2942* -5.3080** 4 LNINF -1.8027-3.3916** -3.9629** 5 LNLTD -9.2524** -9.4327** -9.3718** 6 LNOP -3.6648** -4.5342** -4.5110** 7 LNRES -4.0914** -5.5667** -8.1919** * denotes significance at the level 5% and ** denotes significance at the level 1%. Critical values obtained from Mackinnon (1991) are -2.85, -3.41, and -1.94 at 5% level and -3.43, -3.96, and -2.56 at 1% for first second and third model respectively. 95
Table 6.4: Cointegration Test Results (Trace) (India) Hypothesized Eigenvalue Trace Statistic P values No. of CE(s) None * 0.700581 218.9815 0.0000 At most 1 * 0.455655 130.9501 0.0000 At most 2 * 0.439117 86.55355 0.0013 At most 3 0.264717 44.34179 0.1030 At most 4 0.191001 21.89427 0.3044 At most 5 0.081650 6.421340 0.6457 At most 6 0.002783 0.203478 0.6519 Trace test indicates 3 cointegrating equations at the 0.05 level. *denotes rejection of the hypothesis at the 0.05 level The implication is that there is a possibility to have a co-integrating vector whose coefficient can directly be interpreted as long-term equilibrium. So, Cointegration analysis has been used to study the long run relationship among the variables under study. Firstly Cointegration Trace Test and Maximum Eigenvalue test have been applied to check the cointegration relationship. Results of these tests have been reported in Table 6.3 and in Table 6.4, which shows three cointegrating vectors. This cointegrating relationship represents the foundation of a complete Vector Error Correction Model (VECM). Three alternative cointegrating equations representing the relationships among the variables under study have been obtained after executing cointegration test. However, in these equations FDI appears on the right hand side as an independent variable. As the objective of this study was to detect the determinants of FDI so there was need to impose following restrictions on cointegration relationship. (See annexure I) 96
β11 = 1, β12 = 0, β13 = 0 β21 = 1, β23 = 0, β24 = 0 β31 = 1, β34 = 0, β35 = 0 With these restrictions three alternative equilibrium relationship defining the determinants of FDI have been obtained in order to investigate impact of all the variables on the FDI. (See Harris and Sollis, 2006 for details on imposing restrictions) Table 6.5: Cointegration Test Results (Maximum Eigenvalue) (India) Hypothesized No. of CE(s) Eigenvalue Max-Eigen Statistic P values None * 0.700581 88.03142 0.0000 At most 1 * 0.455655 44.39657 0.0153 At most 2 * 0.439117 42.21176 0.0040 At most 3 0.264717 22.44753 0.1983 At most 4 0.191001 15.47293 0.2571 At most 5 0.081650 6.217861 0.5854 At most 6 0.002783 0.203478 0.6519 Max-eigenvalue test indicates 3 cointegrating equations at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level Table 6.6: Estimated Cointegrating Relationship (India) Equations Independent Variables Impact on FDI 1 LNGDP 5.035** (4.70) 2 LNEXCH -5.351** -(9.92) 3 LNOP 19.832** (5.86) 4 LNRES 10.842** (10.82) 5 LNLTD 6.576** (6.51) 6 LNINF -13.574** -(10.58) ** denotes significance at the level 1%. Figures in Parentheses are t values 97
VECM has been applied to obtain the final results of the analysis which are reported in the Table 6.5. The results of this analysis have also been represented in equation 6.1 given as below: FDI = 13.816+5.035GDP-5.351EXCH+19.832OP+10.842RES+6.576LTD-13.574INF + u t ------- (6.1) (4.702) - (9.92) (5.86) (10.82) (6.51) - (10.58) The analysis shows that combinations of variables like GDP, RES, OP, INF, EXCH, and LTD are found to be statistically significant in India. Here X coefficients (elasticities) show the percentage change in FDI due to one % change in other variables taken in the study. X coefficient of GDP is estimated to be 5.035 found to give positive and statistically significant impact indicating 1% Change in GDP will raise FDI by 5.035%. Larger market size (GDP), faster economic growth and higher degree of economic development provide more and better opportunities for the foreign investors to expand and exploit developed resources of the country for taking all the profitable economic advantages Exchange rate is found to be significant variable having negative impact on FDI as the coefficient of this variable is determined as -5.351. It shows that the 1% change in this variable will tend to decrease 5.351% in FDI. Volatility of exchange rate and frequent change in the value of currency create the uncertainty among the foreign investors about the price stability and monetary regulatory mechanism of the country concerned. It discourages the foreign investors to invest in a country with exchange rate instability. Inflation is also estimated to be statistically significant variable affecting FDI as it explains that the 13.57% variation in FDI due to 1% change in inflation. The value of X coefficient is estimated to be -13.574 depicts the negative impact of inflation on FDI. This is because of the reason that high level of price in the country results in rising cost of production on account of increase in input prices like wages, cost of raw material, land prices and cost of capital. High price of the product also adversely affects domestic as well as international demand of product. All these factors ultimately lead to reduction in profitability in business thus discourage foreign investment in the countries with high inflation rate. 98
X coefficient of 6.576 has been calculated so far as the LTD as a determinant of FDI is concerned showing 1% increase in LTD would bring 6.57% variation in FDI. It also indicates the positive impact of this variable on FDI in case of India that is due to the optimum and extensive utilisation of these funds for different growth oriented economic activities like expansion of means of transport and communication, generation of power, development of banking and financial sector in the economy. This in turn made India to visualise a tremendous growth rate in manufacturing sector in past years that resulted in more FDI inflows in the country. Openness is found to be important variable having positive and significant impact on FDI as the coefficient of this variable is registered as 19.832 which shows the that 1 % change in this variable has tendency to bring 19.83% increase in FDI. It shows that the more an emerging market tries to open its economy to outside external trade, the more it can attract FDI. Export oriented FDI depends upon liberal trade policies reflected in openness of the country as the transnational corporations are not interested in market seeking behaviour initially. The relationship between reserves and FDI is found to be positive and statistically significant. The coefficient of this variable is determined as 10.842 indicating that 1% increase in reserves would cause the FDI to rise by 10.84%. As high level of foreign exchange reserves reflects the strength of external payments position and helps to improve the confidence of the prospective investors. The variable openness, inflation and Reserve have been found to be the major contributor in explaining 19.83%, 13.57% and 10.84% variation respectively in FDI when there is 1% change in these variables. Variables openness, Reserve, GDP, and LTD have estimated to give positive impact to FDI while negative impact of Inflation and Exchange rate has been noticed on FDI. 6.7 Statistical Diagnostic for China The stepwise regression analysis has been applied to find out the impact of variables like GDP, INF, OP, EXCH, LTD and RES on FDI in China. Using Ordinary Least Square, these explanatory variables are regressed to find out determinants of FDI. The regression results of the above said analysis have been reported in Table 6.6 which shows that a combination of variables like LTD, RES and EXCH have been 99
found to be statistically significant in China, while the coefficients of other variables like GDP, INF and OP have shown insignificant t-value. The value of F is found to be significant indicating the significance of the model. The value of adjusted R 2 has been estimated to be 0.921 which indicates the 92 % variation in FDI is explained by the combination of these three variables i.e LTD, RES and EXCH in case of China, but a very low value i.e. 1.194 of Durbin-Watson statistics and such a high value of Adjusted R 2 reveals the existence of spurious regression due to problem of autocorrelation. Moreover, the values of VIF for all the variables have found to be greater than 1 which indicates the interdependence among the explanatory variables suggesting the problem of multicollinearity. There arises a need to verify the reliability and robustness of these regression results by applying the cointegration technique for finding the variables influencing FDI. Table: 6.7 Regression Results (FDI as Dependent Variable) (China) Variables Coefficients Std. Error t-statistics Prob. VIF C -5.310 1.213-4.378 0.000 - LTD 0.580 0.096 6.041 0.000 2.595 RES 0.533 0.167 3.184 0.003 3.568 EXCH 1.253 0.431 2.909 0.007 3.387 Adjusted R 2 0.921 F-statistics 1.016 Prob. (F-statistic) 0.000 Durbin-Watson 1.194 6.8 Econometric Methodology There is a need to check the stationarity of each data series before estimating any relationships between FDI and its explanatory variables. All the variables in this analysis have been taken in logarithmic form to make them stationary at lesser order of integration. The stationarity of these variables has been tested by applying formal 100
Table 6.8: Augmented Dickey Fuller Test Results Unit Root Tests at Logarithmic levels (China) Sr. No. Variables Without Drift and Time Trend With Drift With Drift and Time Trend 1 LNFDI 0.6840-4.1272* -2.1760 2 LNLTD -0.7906-4.7332* -4.0304* 3 LNEXCH 0.2382-1.1759 0.0992 4 LNRES 3.7463-0.6648-3.3241 Unit Root Tests at First Differences Sr. No. Variables Without Drift and Time Trend With Drift With Drift and Time Trend 1 LNFDI -4.1029-3.7463* -4.9632* 2 LNLTD -4.4278* -2.6660* -5.9029** 3 LNEXCH -4.0281* -4.0419** -3.4967* 4 LNRES -1.8907* -6.0382* -5.9065* * denotes significance at the level 5% and ** denotes significance at the level 1%. Critical values obtained from Mackinnon (1991) are -2.85, -3.41, and -1.94 at 5% level and -3.43, -3.96, and -2.56 at 1% for first second and third model respectively. Table 6.9: Cointegration Test Results (Trace) (China) Hypothesized Eigenvalue Trace Statistic P values No. of CE(s) None * 0.990704 212.2737 0.0000 At most 1 * 0.804975 67.25049 0.0000 At most 2 * 0.410514 16.57697 0.0342 At most 3 0.006217 0.193329 0.6602 Trace test indicates 2cointegrating equations at the 0.05 level. *denotes rejection of the hypothesis at the 0.05 level 101
unit root procedure i.e. the Augmented Dickey-Fuller (ADF) test. Cointegration analysis has been applied to study the long run relationship among these variables for estimating impact of these variables on FDI in India. As the quarterly data of these variables was not available for the period under study in case of China, therefore it was not feasible to find the influence of all the six variables on FDI by applying cointegration technique due to lesser number of observations and more number of variables causing the loss of degree of freedom in the analysis. Only three explanatory variables can be examined under cointegration analysis on account of the above mentioned limitation, so there arises the problem of choice among all the independent variables taken in the study. It was decided to choose these three variables i.e LNEXCH, LNLTD, and LNRES on the basis of their significance obtained in the regression results as mentioned in the Table 6.6. 6.9 Results and Discussion The ADF test results for four variables such as LNFDI, LNEXCH, LNLTD, and LNRES have been reported in Table 6.7 which shows that the null hypothesis of presence of unit root has been rejected for all the first difference variables specified. This means that the series are non-stationary at level but stationary at first-differences. Cointegration analysis has been applied to estimate the long run relationship among these variables. Firstly Cointegration Trace Test and Maximum Eigenvalue test have been used to find out the existence of cointegration relationship. Table 6.8 and in Table 6.9, exhibit these results showing the presence of three cointegrating vectors which represents the foundation of a complete Vector Error Correction Model (VECM). Results of which have been reported in the Table 6.10. Table 6.10: Cointegration Test Results (Maximum Eigenvalue) (China) Hypothesized Eigenvalue Max-Eigen Statistic P values No. of CE(s) None * 0.990704 145.0232 0.0000 At most 1 * 0.804975 50.67352 0.0000 At most 2 * 0.410514 16.38365 0.0228 At most 3 0.006217 0.193329 0.6602 Max-eigenvalue test indicates 2cointegrating equations at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level 102
There have been obtained three alternative cointegrating equations after applying cointegration test representing the relationships among the variables under study. However, in these equations FDI appears on the right hand side as an independent variable but the objective of this analysis was to identify the determinants of FDI in China. Therefore some restrictions on cointegration relationship are required to be imposed. (See annexure I) which are given as under. β11 = 1, β13 = 0, β14 = 0 β21 = 1, β22 = 0, β24 = 0 β31 = 1, β32= 0, β33 = 0 With these restrictions three alternative equilibrium relationships defining the determinants of FDI have been obtained in order to investigate impact of all the variables on the FDI. VECM has been executed to obtain the results of the analysis which are shown in the Table 6.10. These results have also been expressed with the help of equation 6.2 given as under: FDI = -5.6502 +1.874 EXCH + 1.502LTD + 6.836RES ------- (6.2) (11.10) (12.79) (6.06) Table 6.11: Estimated Cointegrating Relationship (China) Equations Independent Variables Impact on FDI 1 LNEXCH 1.874** (11.10) 2 LNLTD 1.502** (12.79) 3 LNRES 6.836** (6.06) ** denotes significance at the level 1%. Figures in Parentheses are t values The results of the analysis reveal that all the variables such as, EXCH, LTD and RES have found to be statistically significant in China. X coefficients (elasticities) of 103
these variables depict the percentage change in FDI due to one % change in these variables taken in the study. X coefficient of exchange rate is estimated to be 1.874 indicating positive and statistically significant impact of this variable on FDI. 1% Change in EXCH has been found to change FDI by 1.874%. The relationship between relative exchange rate and FDI can be attributed to the fact that the appreciation of the source country currency relative to that of the host country currency will reduce the relative cost of capital and enable MNCs to invest more in that country as compared to countries with depreciated currency (Liu, 2010). Long term debt is determined to be a significant variable causing positive impact on FDI as the coefficient of this variable is evaluated as 1.502 which shows that the 1% increase in this variable will tend to increase FDI by 1.502%. Effective utilisation of these funds for different growth oriented economic activities like expansion of means of transport and communication, generation of power, development of banking and financial sector in the economy resulted in more FDI inflows in the country The variable Reserves is found to be important variable having positive and significant impact on FDI as the coefficient of this variable has been calculated as 6.836 which shows the that 1 % change in this variable has tendency to bring 6.836 % increase in FDI. As high level of foreign exchange reserves reflects the strength of external payments position and helps to improve the confidence of the prospective investors. All the variables like Exchange rate, Long term debt and Foreign exchange reserves have been examined to be give positive impact to FDI and Foreign exchange reserves has found to be the most important variable causing more variation in FDI as compared to other explanatory variables analysed in the study. In brief it can be said that X coefficients have shown positive and significant impact of Exchange rate, Long term debt and Foreign exchange reserves in case of China. In India, the variable Gross domestic product (GDP), Foreign exchange reserves (RES), Openness (OP i.e sum of Exports and Imports as a percentage of GDP) and Long Term Debt (LTD) has shown positive influence on FDI. However the 104
Exchange rate (EXCH) and Inflation (INF) has negative impact on FDI. Conflicting results of the exchange rate in both the countries can be attributed to the fact that If FDI aim at producing for re-exports, it is complementary to the international trade. Thus an appreciation of the local currency is supposed to reduce the FDI inflows since it raises the local labour costs. A decrease in the relative labour cost, either through a fall in its relative wages or real exchange rate deprecation, will increase the foreign investment. On the other hand, if FDI aim at serving the local market, FDI and trade are substitutes of each other. An appreciation of the local currency increase FDI inflows due to higher purchasing power of the local consumers. The depreciation in the real exchange rate of the FDI recipient countries will increase the FDI inflow since it reduced cost of capital investment (Ren and Pentecost, 2008). The results of this analysis substantiate the findings of some previous studies. Cheng & Ma, (2008) observed positive and significant impact of real GDP, real per capita GDP, foreign reserves and currency appreciation on the FDI flows. Shan (2002) found that Labour supply, low Labour wage, Exports and Exchange rate has significant and positive influence on FDI inflows. Dees (1998) concluded that the variables Market size, low labour wage and Exchange rate have positive impact on FDI. Botric and Skuflic (2005) and Casi and Resmini (2010) have found positive influence of GDP and openness on FDI inflows in South East European countries and some EU regions. Venkataramany (2002) found that the variables exports, GDP, Terms of Trade have positive and Inflation has negative but significant impact on FDI. Banga (2003) estimated the positive impact of GDP, Education and external debt on FDI while exchange rate has been found to influence it negatively. Helldin (2007) observed the positive and significant impact of GDP and domestic investment and negative impact of Exchange rate on FDI. However, the determinants of FDI differ from country to country depending upon other incentives available in the country. The economic parameters have to be in order as well, so that Indian economy can compete with China with more competitive strengths to increase its share in global foreign direct investments. The substantial amount of foreign direct investment from all over the world has played an important role in the growth of the economy. 105