T h e C a s e o f Tu r k e y

JÖ N K Ö P I N G I N T E R N A T I O N A L BU S I N E S S S C H O O L JÖ N KÖ PIN G UN IVERSITY FDI and GROWTH T h e C a s e o f Tu r k e y Bachelor Thesis within Economics Author: Bengü Kaya Supervisor: Per-Olof Bjuggren Deputy supervisor: Andreas Högberg Jönköping June 2009

Acknowledgements I would like to thank my supervisor Per-Olof Bjuggren and deputy supervisor Andreas Högberg for the help, guidance and support during the process of writing this thesis. I would like to thank especially to my parents, my brother, my boyfriend and my daughter Asya for their great patience. Without their support, love and encouragement I would not finish my Bachelor thesis. June, 2009, Jönköping, Sweden Bengü Kaya i

Foreign Direct Investment, Economic Growth, Eclectic Paradigm, Turkey, Granger Causality, Time Series JEL Classification Abstract F21, F43, 053, C01, C22 ii

List of abbreviations iii

Table of Contents 1 Introduction... 1 2 Background... 3 2.1 Turkey s Economy... 3 2.2 Overview of FDI in Turkey... 3 2.3 Previous Studies... 6 3 Theoretical Framework... 8 3.1 Definition of FDI... 8 3.2 Different Types of FDI... 8 3.3 Benefits and Disadvantages of FDI... 9 3.4 Determinants of FDI... 10 3.5 FDI Theories... 11 3.5.1 The Evolution of FDI Theories... 11 3.5.2 The Eclectic Paradigm... 11 3.5.2.1 Ownership Advantage (O)... 11 3.5.2.2 Location Advantage (L)... 12 3.5.2.3 Internalization Advantage (I)... 12 3.6 Location Advantages of Turkey... 13 4 Method... 14 5 Empirical Analysis... 16 5.1 Descriptive Data... 16 5.2 Granger Causality... 16 5.2.1 Data Statistics... 17 5.2.2 Unit root test... 17 5.2.3 Granger Causality... 17 5.3 Regression Results and Analysis... 18 6 Conclusion and Suggestions for Further Research... 21 7 References... 22 8 Appendices... 27 8.1 Determinants of international products... 27 8.2 Summary of Previous Studies... 28 8.3 Turkey s location advantages for FDI... 29 8.4 Correlation Matrix of FDI and Growth... 29 8.5 FDI Inflows % and GDP %, 1970-2007... 30 8.6 Descriptive statistics of Granger Causality... 30 8.7 D (FDI) Correlograms Table... 30 8.8 D (GDP) Correlograms Table... 31 8.9 Level and First differences of FDI and GDP... 31 8.10 Descriptive Statistics of Regression Model Variables... 32 8.11 Correlation Matrix of Estimated Model Variables... 32 8.12 Summary of Variables... 33 iv

8.13 Population Distribution in Turkey... 33 Figure 1: FDI Inflows to Turkey and World... 1 Table 1: Augmented Dickey-Fuller Test... 17 Table 2: Unrestricted Cointegration Tests... 17 Table 3: Pairwise Granger Causality Tests... 18 Table 4: Sample Regression Results of Estimated FDI Model... 20 v

1 1 Introduction Figure 1: FDI Inflows to Turkey and World 2.40E+10 2.00E+12 2.00E+10 1.60E+12 1.60E+10 1.20E+12 1.20E+10 8.00E+11 8.00E+09 fdiw 4.00E+11 4.00E+09 FDI 0.00E+00 1970 1975 1980 1985 1990 1995 2000 2005 Source, WID (2008), where fdiw represents FDI inflows in world and fdi is representing inflows in Turkey.

2

3 2 Background This section is divided into three parts. The first part presents Turkey s economic history starting from 1960s and continuing until the present time. The effect of FDI in Turkey is introduced in the second part. And in the last part the previous empirical studies are presented. 2.1 Turkey s Economy 2.2 Overview of FDI in Turkey

4

5

6 2.3 Previous Studies

7

8 3 Theoretical Framework This section is divided into six parts. First part gives brief different definitions of FDI. In the second part types of FDI are introduced. The following part presents advantages and disadvantages of the FDI inflows by the view of MNEs and host country. The fourth part presents the theories of FDI and growth. The most common used FDI theory, eclectic paradigm, is presented in the fifth part with its specific advantage stages. And finally in the sixth part location advantages of Turkey is introduced. 3.1 Definition of FDI 3.2 Different Types of FDI

9 3.3 Benefits and Disadvantages of FDI

10 3.4 Determinants of FDI

11 3.5 FDI Theories 3.5.1 The Evolution of FDI Theories 3.5.2 The Eclectic Paradigm 3.5.2.1 Ownership Advantage (O)

12 ö 3.5.2.2 Location Advantage (L) 3.5.2.3 Internalization Advantage (I) 1 For the readers strategic determinants of international products are summarized in Appendix 8.1

13 3.6 Location Advantages of Turkey

14 4 Method 2 The explanatory variables did not give any significant results on the dependent variable. Due to getting the best result all variables are lagged one year.

15 3 Summary of regression variables are in Appendix 8.12.

16 5 Empirical Analysis In the first part of this section descriptive data of GDP growth and FDI inflows are presented. The log-run and short-run relationship of economic growth and FDI is analyzed in the second part. And finally in the last part the regression model analysis is presented. 5.1 Descriptive Data 5.2 Granger Causality

17 5.2.1 Data Statistics 5.2.2 Unit root test Table 1: Augmented Dickey-Fuller Test Variable t-statistic Probability(5% level) Growth -6.860497-2.948404 FDI -3.705381-2.951125 Table 2: Unrestricted Cointegration Tests Hypothesis Trace Statistic 5% critical value Probability None* 16.65372 15.49471 0.0333 At most 1* 1.395555 3.841466 0.2375 *denotes rejection of the hypothesis at 5% level 5.2.3 Granger Causality 4 All the steps of ADF test are introduced in Appendices 8.7, 8.8 and 8.9.

18 Table 3: Pairwise Granger Causality Tests Hypothesis F-Statistics Probability GROWTH does not Granger Cause FDI 4.27135 0.04644 FDI does not Granger Cause GROWTH 0.22509 0.63823 5.3 Regression Results and Analysis 5 For the readers correlation matrix of estimated model variables is introduced in Appendix 8.11.

19 6 Monthly minimum wage is almost $417 (666TL) for each worker in 2009 (source author, according to her knowledge as a Turkish citizen).

20 Variable 1 2 3 4 5 6 GROWTH t-1 2.97E+08 (1.48) 2.31E+08 (1.45) 2.41E+08 (1.58) 76424570 (0.52) 1.14E+08 (1.71)* 1.11E+08 (1.70)* OPEN t-1 2.17E+10 (4.71)*** 8.99E+09 (1.18) 1.02E+10 (1.51) -1.27E+09 (-0.39) POP t-1-4.98e+09 (-2.05)*** -4.80E+09 (-2.21)** -3.04E+09 (-3.03)*** -2.76E+09 (-3.98)*** INFL t-1-65459375 (-3.07)*** -20010950 (-1.90)* -20731002 (-2.02)* FDI t-1 1.042268 (11.09)*** 1.030671 (11.73)*** R 2 0.059010 0.431012 0.495159 0.609888 0.921540 0.921157 Adj.R 2 0.032125 0.397542 0.449262 0.561124 0.908885 0.911302 N 37 37 37 37 37 37 DW Stat. 1.859655 1.837661 *, **, *** represents 10, 5 and 1 percent level

21 6 Conclusion and Suggestions for Further Research

22 7 References Berkoz, L. & Turk, S.S (2007). Yabancı Yatırımların Yerseçimini Etkileyen Faktörler: Türkiye örneği, from http://www.itudergi.itu.edu.tr/tammetin/itu-a_2007_6_2_l_berkoz.pdf

23 Doğrudan ı ö

24 http://www.fdimagazine.com/news/fullstory.php/aid/1266/eu_entry_poses_threat_to_f ree_zones.html http://www.econ.ku.dk/research/publications/pink/2004/0430.pdf ö ö

25 TUIK (2008). Statistics Data http://www.turkstat.gov.tr/altkategori.do?ust_id=4 Turkish Embassy (2007). Recent Economic Developments and Buying Property in Turkey, presentation paper written by Aziz Dogan on June, 5, 2007, from http://www.turkisheconomy.org.uk/tbcci_presentation.pdf

26 Türkiye için Doğrudan Yabancı Yatırım Stratejisi ne Doğru, from http://www.iaw.unibremen.de/~jtholen/tagungen/papers/yilmaz.pdf

27 8 Appendices 8.1 Determinants of international products Types of international production Ownership advantages Location advantages Internalization advantages Strategic goals of MNEs Types of activity that favor MNEs re- seek- Natural source ing Capital, technology, access to markets. Possession of natural resources. To ensure stability of supplies at right price; control markets. To gain privileged access to resources vis-avis competitors. Oil, copper, bauxite, bananas, hotels. Market seeking Capital, technology, management and organizational skills; surplus R&D and other capacity; economies of scale; trademarks. Material and labor costs; market size; government policy. Wish to reduce high transaction or information costs, buyer uncertainty, etc; to protect property rights. To protect existing markets, counteract behavior of competitors; to preclude rivals from gaining new markets. Computers, pharmaceuticals, motor vehicles, cigarettes Efficiency seeking As above but also access to markets. Low labor costs incentives to local production by host governments. The economies of vertical integration As part of regional or global product rationalization and to gain advantages of process specialization. Consumer electronics, textiles and clothing, cameras, etc. Strategic asset seeking Any of first three that offer opportunities for synergy with existing assets Any of first three that offer technology, markets and other assets in which firm is deficient Economies of common governance; improved competitive or strategic advantage; to reduce risks. To strengthen global innovatory or production competitiveness; to gain new product lines or markets. Industries that record a high ratio of fixed to overhead costs and which offer substantial economies of scale. Adopted from Dunning 1980, 1993

28 8.2 Summary of Previous Studies Author Data Method Result Balasubramanyam et al. (1996) Cross-section data,46 developing countries between 1970-1985 OLS regressions. FDI has positive effect on export supported countries rather than import substituting Borensztein et al. (1998) Cross-section data, 69 countries between 1970-1989 Regression analysis. FDI and growth are positively related if the host country has high level of human capital Alfaro et al. (2004) Panel data, data set of credit market and equity market indicators in OECD and non- OECD countries between 1975-1995 OLS regression. FDI affects the growth positively, if the country has good financial market Chowdhury and Mavrotas (2006) Time series, Chile, Malaysia and Thailand between 1969-2000 Toda and Yamamoto causality test FDI and growth affect each other only in Chile; other two countries have only bidirectional causality. Johnson (2005) Cross-section and panel data. 90 developing and developed countries between 1990 and 2002 OLS regression and REM FDI inflow and domestic investment have positive effects on economic growth in developing countries but not in developed ones. FDI is affected by corruption negatively in developing countries. Hansen and Rand (2006) Panel data, 31 countries between 1970-2000 Granger causality test If the gross capital of FDI increases, the growth will increase too. Pramadhani et al. (2007) Cross-section data, Indonesia between 1994-2000 Granger causality test FDI and exports are positively related. Busse Groizard (2008) and Panel data, 84 countries between 1994-2003 Cross-country growth regression, generalized method of moments (GMM) Growth and FDI have limited relation by government regulations. Carkovic and Levine (2002) Panel and crosssection data, 72 countries between 1960-1995 OLS and GMM FDI inflow has no effect on growth.

29 Dutt (1997) Cross-country data, North- South countries between 1985-1994 GLS There is no significant relationship between FDI and growth. Basu et al. (2003) Panel data, selected 23 developing countries, between 1990-1999 Two-way causality test In closed economy the short-run causality is bidirectional but in the longrun it is from growth to FDI. De (1997) Mello Survey Analysis The direction of causality between FDI and economic growth depend on the impacts of FDI determinants. De (1999) Mello Cross section data, 32 OECD and non- OECD Causality test In the non-oecd countries FDI and growth have no significant relationship. 8.3 Turkey s location advantages for FDI rights Key location factors Market seeking FDI Economic size Economic growth Population size Per capita incomes Efficiency seeking FDI Labor cost Labor productivity Regional integration zone Labor skills and supply Asset seeking FDI Supply of engineers and technicians R&D and innovation base FDI enabling environment Independent FDI Privatization FDI Facilitation process Political commitment Incentives Investment promotion Institutional-Political environment Economic stability Policy certainty Political interference and corruption Justice system and intellectual property Internal social tensions Competitive position Strong Strong Strong Medium Strong Strong Strong Strong Strong Weak Strong Weak Medium Weak Strong Weak Weak Weak Weak Weak Weak Adopted from Loewendahl & E-Loewendahl (2001) 8.4 Correlation Matrix of FDI and Growth Correlation FDI GROWTH FDI 1-0.446885 GROWTH -0.446885 1

30 8.5 FDI Inflows % and GDP %, 1970-2007 30 gdp% 20 10 0-10 -20-30 -40 1970 1975 1980 1985 1990 1995 2000 2005 500 FDI% 400 300 200 100 0-100 1970 1975 1980 1985 1990 1995 2000 2005 8.6 Descriptive statistics of Granger Causality Variable Mean Maximum Minimum Std. Deviation GROWTH 0.037491 12.54641-12.47193 5.809606 FDI 5.94E+08 9.96E+09-2.22E+09 2.08E+09 8.7 D (FDI) Correlograms Table Autocorrelation Partial Correlation AC PAC Q-Stat Prob. *.. *. 1-0.161-0.161 1.0366 0.309. *.. *. 2-0.068-0.096 1.2251 0.542.**..**. 3-0.226-0.262 3.3858 0.336. *... 4 0.128 0.035 4.1065 0.392. **.. **. 5 0.197 0.199 5.8574 0.320.... 6-0.040-0.004 5.9301 0.431... *. 7 0.003 0.082 5.9306 0.548.... 8-0.053 0.044 6.0730 0.639

31 8.8 D (GDP) Correlograms Table Autocorrelation Partial Correlation AC PAC Q-Stat Prob ****. ****. 1-0.548-0.548 12.038 0.001. *..**. 2 0.095-0.293 12.410 0.002. *... 3 0.094 0.002 12.788 0.005.**..**. 4-0.281-0.289 16.230 0.003. **.. *. 5 0.225-0.131 18.517 0.002. *..**. 6-0.175-0.229 19.940 0.003. *... 7 0.139-0.046 20.872 0.004. *..**. 8-0.146-0.288 21.934 0.005 8.9 Level and First differences of FDI and GDP 1.20E+10 dfdi 1.20E+10 15 dgdpgrowth 15 1.00E+10 8.00E+09 6.00E+09 1.00E+10 8.00E+09 6.00E+09 10 5 10 5 4.00E+09 4.00E+09 0 0 2.00E+09 0.00E+00-2.00E+09 2.00E+09 0.00E+00-2.00E+09-5 -10-5 -10-4.00E+09 1970 1975 1980 1985 1990 1995 2000 2005-4.00E+09-15 1970 1975 1980 1985 1990 1995 2000 2005-15 2.40E+10 FDI 2.40E+10 12 gdpgrowth 12 2.00E+10 2.00E+10 8 8 1.60E+10 1.60E+10 4 4 1.20E+10 1.20E+10 8.00E+09 8.00E+09 0 0 4.00E+09 4.00E+09-4 -4 0.00E+00 1970 1975 1980 1985 1990 1995 2000 2005 0.00E+00-8 1970 1975 1980 1985 1990 1995 2000 2005-8

32 8.10 Descriptive Statistics of Regression Model Variables FDI GROWTH OPEN POP INFL Mean 1.89E+09 4.462564 0.284083 2.009154 46.23112 Median 4.81E+08 5.150450 0.252861 2.068482 47.89100 Maximum 2.20E+10 10.46118 0.568831 2.593257 103.0160 Minimum 10000000-5.697477 0.083215 0.617152 7.084290 Std. Dev. 4.88E+09 3.977765 0.145387 0.446993 27.84345 Skewness 3.423021-0.895921 0.444170-0.909220 0.228418 Kurtosis 13.54841 3.203863 2.127382 3.821983 1.914954 Jarque-Bera 250.3838 5.149409 2.455135 6.305432 2.194539 Probability 0.000000 0.076176 0.293005 0.042736 0.333781 Sum 7.20E+10 169.5774 10.79515 76.34784 1756.782 Sum Sq. Dev. 8.80E+20 585.4368 0.782083 7.392715 28684.54 Observations 38 38 38 38 38 8.11 Correlation Matrix of Estimated Model Variables FDI GROWTH OPEN POP INFL FDI 1 0.098791741 0.624087039-0.576680624-0.374071302 GROWTH 0.098791741 1 0.085678947-0.053968863-0.354011396 OPEN 0.624087039 0.085678947 1-0.832385391-0.050119850 POP -0.576680624-0.053968863-0.832385391 1 0.052812134 INFL -0.374071302-0.354011396-0.050119850 0.052812134 1

33 8.12 Summary of Variables Variable Explanation Data Source, Period Expected sign Dependent variable Explanatory variables 8.13 Population Distribution in Turkey