Does China Follow Japan & South Korea?
Gravity Model vs. Flying Geese Theory
MSc International Financial Management Faculty of Economics and Business
University of Groningen
Master Thesis Supervisor: Dr. H. Vrolijk Co-assesser: Dr. W. Westerman
Student Name: Di Liu Student Number: s2499029
Content
ABSTRACT... 2
1. Introduction... 3
2. Literature Review...5
2.1 Gravity Model...5
2.2 Flying Geese Theory...6
2.3 OFDI in Japan...10
2.4 Empirical studies on Gravity Model and Flying Geese Theory...12
2.5 Research Hypothesis...12
3. Methodology Design...14
3.1 Base Gravity Model...15
3.2 Research Model...16
3.3 Sample Description...17
4. Results and Discussions...18
4.1 Results... 18
4.2 Discussions...23
4.2.1 Gravity model...23
4.2.2 Flying geese model...24
4.2.3 Control variables...25
4.2.4 China’s outward FDI pattern...27
5. Conclusions... 29
Acknowledgement... 31
Appendix 1... 31
Appendix 2... 33
ABSTRACT
The Gravity model is often used to test the determinants of outward foreign direct investment (OFDI). The Flying Geese Theory is first introduced to depict the economic development pattern and OFDI-pattern of Japan in East Asia, but now can be generalised to a wider area. In this thesis both models are applied to Japan, South Korea and China, the largest developing economy in the world that not only have large inward FDI also small but growing outward FDI.
This thesis shows that GDP has a positive, and distance a negative impact on OFDI of Japan, South Korea, and China. This is a confirmation of the gravidty model. For testing the flying geese model GDP per capita is used as a crude indicator for relative labor costs. For the OFDI to developed countries I found a positive relationship, as a higher income level is attractive for foreign investors. However, GDP per capita has a negative relationship with the OFDI to developing countries, as predicted by the flying geese model. Both gravity model and flying geese model can explain international investment in the East-Asian region, but the OFDI-pattern of China fits less.
1. Introduction
The foreign direct investment (FDI) is a popular issue that has drawn much attention. OECD (FDI in Figures, 2014) defines the FDI as “A category of investment that
reflects the objective of establishing a lasting interest by a resident enterprise in one economy (direct investor) in an enterprise (direct investment enterprise) that is resident in an economy other than that of the direct investor.” The outward
investments are made by direct investors who resident in the reporting country that invest in cross-border investments, whilst the inward investments are made by non-resident ones in the reporting country. In general, FDI flows are financial transactions made across the border within a given period between subsidiaries that are in a direct investment relationship; FDI positions are in regard to the stock of investments at a given point in time; and FDI income is the return on direct investment positions of equity and debt.
By definition, inward FDI should be equal to the outward FDI globally despite of statistical discrepancies. Unexpectedly, although there appeared so many previous studies of inward foreign direct investment (IFDI), the existing research on outward foreign investment (OFDI) is still limited. Especially for the fast developing economy China - the largest emerging economy, even though its OFDI has grown significantly in recent decade of years, this maintains quite less compared to its IFDI.
Flying Geese pattern, which is first proposed by Akumasu and then is further elaborated and expanded by Kojima (1960). The flying geese pattern not only can describe the original economic development or trade, but also can depict the foreign direct investment topic.
According to World Economic Outlook (IMF, 2014), Japan and South Korea are of the advanced economies of East Asia. For example, the GDP per capita (World Bank, 2013) of Japan and South Korea in 2013 are US$36,315 (ranks 25th) and US$33,140 (ranks 30th), respectively. Moreover, The Human Development Report (UN, 2014) shows that the estimated yearly Human Development Index (HDI) ranking of Japan is 17th (0.890), while South Korea ranks even higher as 15th (0.891). This HDI result shows that both countries already have reached a high standard of life expectancy, education, and income indices. Along with these high rankings is both countries’ FDI.
When it comes to the outward FDI, China’s OFDI is still relatively so small compared to its inward FDI, even compared to some developing countries. What is more, China’s OFDI is mainly composed of acquisitions in East Asia and resource-rich regions of Africa (Morck et al., 2007). The question is whether China will follow the same OFDI-pattern as its two close neighbour countries Japan and South Korea.
From a country level, this topic OFDI-pattern in flying geese model can be a reflection of a country’s financial, managerial and organisational transformation, it can help with industrial upgrading, and it is also a connection to the world’s trend. From a firm level, this topic can help a firm especially a multinational corporation to set a strategic management planning to achieve better business performance.
2. Literature Review
Two theoretical models are most relevant to find determinants and patterns in outward FDI: the Gravity Model and the Flying Geese Theory.
2.1 Gravity Model
Anderson (2010) claims that the gravity model is one of the most successful empirical models in economic interaction across space between trade and factor movements. Van Bergeijk and Brakman (2010) hold that the gravity model is considered to depict one of the most stable economic relationships as the interaction between large economic clusters is much stronger than that between smaller clusters, and the attraction between nearby clusters is more than remote ones. It is possibly most famous for its applications in international trade and capital flows among countries.
models, so it is not possible to use the gravity model to differentiate from one theory to another.
Taking these disparate views into account, the gravity model is widely used not only in testing trade, but also in testing FDI. Petri (2010) uses it to exploit whether Asian FDI patterns distinguish much from anywhere else, and in what ways if different. The results indicate that Asian FDI flows are in favour of hosts with relatively low technology achievement and strong intellectual property right regimes. Hattari and Rajan (2008) use the augmented gravity model for the determinants of FDI flows, and Asia fit the data well. The determinants of FDI flows from OECD to Asia are different from that of inter-regional flows; especially the elasticity of distance, which is larger for FDI from non-Asia-Pacific OECD countries than FDI for intra-developing Asian flows. But this distance effect fades away when taking different time zones into counted.
2.2 Flying Geese Theory
The flying geese theory is first introduced by a Japanese scholar Kaname Akamatsu in the 1930s. He uses the name flying geese to depict the shape of production, import, and export growth curves exhibited by some of Japan’s modern industries before the World War II (Dowling & Cheang, 2000). Kumagai (2008) illustrates the following three figures to explain Akamatsu’s and Okita’s flying geese pattern.
Figure 2.1 Akamatsu’s Fundamental Flying Geese Pattern of Economic Development
Source: Kumagai (2008)
Figure 2.2 is the flying fish diagram. Akamatsu (1935) depicts the industrial development by expanding the pattern to the “one country – multi-product model”. This figure is also consistent with the fourth stage in figure 2.1. The vertical axis of figure 2.1 represents the value of import, production, and export; while the vertical axis in figure 2.2 represents the net export ratio (NER)1 of goods. In the light of the difference in the vertical axis, Akamatsu finds the sequential patterns within and across industries in economic development.
Figure 2.2 A country’s Flying Fish Diagram of Industrial Development
Source: Kumagai (2008)
Figure 2.3 Okita’s Flying Geese Pattern of Economic Development in East Asia
Source: Kumagai (2008)
Noteworthy is a fraction of discussions turned to be the criticisms towards the flying geese model. For example, Bernard and Ravenhill (1995) blame that the flying geese model results dependence on borrowed technology, capital, etc. then discouraged indigenous innovation. They also argue that the flying geese causes “triangle trade” in Asia that brings Asian countries (except the leader Japan) and the US more deficits. Petri (1988) holds that the industry and export structures became similar among the leading goose and the follower ones, which create dilemmas e.g. overproduction, crucial competitions and protectionism.
FDI has been applied to tap the rich resources in the less developed regions and to recycle its comparative advantage.
Qu et al. (2013) use national data from above-scale manufacturing firms to exam whether the flying geese model can explain the capital migrating from coastal areas to the inland areas in China. They found that this migration had been occurring since the mid-2000s, corresponding to the time of the Lewis turning point2. With investment flocking inward, inland wage rates also catch up with the coastal areas. When the equality in wages is realised, China would substantially relocate labour-intensive industries to other developing regions with cheaper labour wages (e.g. Bangladesh).
2.3 OFDI in Japan
Jun et al. (1993) investigates Japan’s recent trends and determinants in FDI. They find that Asia receives most of Japan’s OFDI in developing countries, which occupies nearly half of the total during 1951-1991. Japan’s investors promptly become active in neighbor countries since the 1970s as a result of the increasing labour costs in Japan that weakened the competitiveness in its own manufacturing. Then a slowdown came in the early 1980s and a rapid growth arose in the second half of 1980s. According to Ministry of Finance, Japan’s outward FDI to all the world’s developing countries is US$ 11,623 million, and to the developing Asian countries is US$ 7,054 million in 1990; hence Japan’s OFDI share in developing Asian countries to total developing countries surpass 60% in 1990 (see Table 2.1). It is obviously that Japan’s main outward investments are within East Asia in the 1980s, which complies the flying geese pattern.
Table 2.1 Japan’s OFDI to Developing Asian countries, 1951-1991
Year Total Asia NIEs ASEAN 4 China Others US$ million US$ % US$ % US$ % US$ % 1951-1957 1972-1980 1981-1985 1986-1991 1986 1987 1988 1989 1990 1991 1951-1991 987 8,844 9,632 33,992 2,327 4,868 5,569 8,238 7,054 5,936 53,455 274 3,263 4,117 17,834 1,531 2,581 3,264 4,901 3,354 2,203 25,488 27.8 36.9 42.7 52.5 65.8 53.0 58.6 59.5 47.5 37.1 47.7 598 5,486 5115 12,635 555 1,031 1,966 2,782 3,236 3,083 23,852 60.6 62.0 53.1 37.2 23.9 21.2 35.3 33.8 45.9 51.9 44.6 0 26 261 3,114 226 1,226 296 438 349 579 3,401 0 0.3 2.7 9.2 9.7 25.2 5.3 5.3 4.9 9.8 6.4 115 69 139 391 15 30 43 117 115 71 714 11.7 0.8 1.4 1.2 0.6 0.6 0.8 1.4 1.6 1.2 1.3
Source: Ministry of Finance, Japan
2.4 Empirical studies on Gravity Model and Flying Geese Theory
Petri (2010) holds that the intra-Asian FDI flows mainly go from high-tech economies to medium-tech economies; in contrast FDI in other parts of the world mainly flows between high-tech economies. He uses a gravity model to test Asian FDI flows pattern and the results show consistence with the flying geese theory that Asian development is the result of technological progress on interregional industrial development from the more advanced economies to these less advanced ones. Geda and Meskel (2007) use gravity model and flying geese model to analyse China and India growth surge to Africa, and they find that the gravity model (econometric estimation of parameters) and the flying geese theory (a non-parametric test) provides this similar evidence, that is there is a significant shift of comparative advantage from the Asian drivers to Africa.
Bano et al. (2013) apply the flying geese theory to Asia to explain the patterns of trade, and use the gravity model to identify important determinants of trade. They identify Japan’s leadership role in providing high-tech inputs in labour-intensive manufacturing in less advanced economies in Asia. Kim and Hirata (2009) use the gravity model to examine the determinants of Japan’s OFDI into higher value-added industries in Asia in 1985-2005, and it confirms flying geese theory that Japan’s FDI to the follower geese’s higher value-added industries relies on the industrialization stage or per capita income level of the host countries.
From these studies, gravity model and flying geese theory have relationships in foreign direct investment and trade. What is more, the regression result of gravity model is generally consistent with flying geese theory.
2.5 Research Hypothesis
Hypothesis 1: Home country's FDI outflows are positively related to host countries' GDP.
According to the gravity model, home country favours large markets, more OFDI will be invested in host countries with high GDP.
Hypothesis 2: Distance between home country and host country has negative effect on outward FDI, and this effect will become smaller.
Cheng & Ma (2010) use the gravity model to test whether distance is significant or not in outward FDI, and it shows significance. Because the shorter the distance is, the less the transportation costs. Thus more outward FDI goes to nearer countries. The flying geese theory says that investment will start in the closer countries, but when labour cost increases, the FDI-flows shift to cheap countries further away.
Hypothesis 3: Depreciation of the host country exchange rate has a positive effect on home country's outward investment.
The effect of exchange rate on FDI has been tested in previous studies Kim and Rhe (2009) apply it into the gravity model. Froot and Stein (1991) use the US FDI data to prove that host country’s depreciation increases home country’s wealth and benefit more, so appreciation in host country has a positive impact on home country’s FDI outflows. The exchange rate can influence the production costs, in the sense that depreciation of the host country exchange rate makes it more attractive to shift production from home country to host country.
Hypothesis 4: Higher inflation rate has a negative effect on home country's outward investment.
The effect of inflation rate on FDI also has been tested in previous study. Kim and Rhe (2009) apply it into the gravity model and has proved significant. Buckley et al. (2007) maintain that inflation rate is an indicator of macroeconomic instability’s quality. He has proved that a higher inflation rate in host country causes a decrease in the real earnings for MNCs from home country, thus inflation will influence the outward FDI to this area. The inflation rate can also influence the production costs in the flying geese theory.
taxation. Chiristians (2005) treats double taxation treaties as a commitment to a favourable overseas investment environment. Di Giovanni (2005) finds that host country’s taxes influence FDI flows significantly. Higher taxes of host country negatively affect the home country’s outward FDI. The purpose of a tax treaty is to decrease the tax flows between countries. Baker (2012) notes for developed countries, having tax treaties with developing countries can prevent tax evasion and can shift revenue from developing countries to themselves; for developing countries, tax treaties can attract more inward FDI from developed countries and can benefit knowledge spillovers and increase tax base.
Hypothesis 6: GDP per capita is negatively related with OFDI in developing countries, but is positively related in developed countries.
According to the flying geese theory, GDP per capita can reflect the development level of a country. GDP per capita is usually low in developing countries, which can display a low labour income, so is the low labour cost, therefore more OFDI is attracted from overseas. GDP per capita is usually high in developed countries, in such countries people have higher income level, which is attractive to foreign investments because investors can seek broader market in high GDP per capita countries.
3. Methodology Design
3.1 Base Gravity Model
Cheng and Ma (2010) used the following gravity equation to estimate the determinants of China’s OFDI:
log (FDIi,t) = α + β1 * log(GDPi,t) + β2 * log (PGDPi,t) + β3 * log (disti) + β4 * ChineseLangi + β5 * Borderi + β6 * Landlocki + β7 * Islandi + β8 * Dummyt (1)
Similar gravity equation had been used to explain the determinants of South Korea’s OFDI by Kim and Rhe (2009):
Yit = β0 + β1 (GDP) it + β2 *(GDPpercapita) it + β3 *(population) it + β4 *(patents)it + β5 *(wages) it + β6*Control (exchangerate) it + β7*Control (inflationrate) it + β8*Dummy (developed country) it + μit (2)
Table 3.1 Comparison of variables both gravity models use
Variable Cheng and Ma model Kim and Rhe model OFDI China’s annual outward FDI flows to
country i in year t
South Korea’s annual outward FDI flows to country i in year t
GDP each country’s year-end GDP each country's year-end GDP GDP per
capita
each country’s year-end GDP per capita
each country's year-end GDP per capita distance distance from the economy’s capital to
Beijing
n.a. Chinese
language
dummy variable whether the economy uses Chinese language or not
n.a. boarder whether the host economy shares a
common border with China
n.a. landlock a landlocked economy n.a.
island an island economy n.a.
population n.a. each country's year-end population patents n.a. each country's number of
annually-applied-for patents
wages n.a. each country's average hourly wage for manufacturing workers Exchange
rate
n.a. each country's annual average exchange rate
Inflation rate n.a. each county's annual average inflation rate Developed n.a. dummy variable that if the country is
3.2 Research Model
OFDI contains outward flows, here to examine the determinants of OFDI flows, I adapt the base equation (1) and (2) to an extended regression model (3):
logYit = α+ β1*log(GDP) it + β2*log(GDP per Capita) it + β3*log(Distance) i + β4* (Exchange Rate) it + β5* (Inflation Rate) it
+ β6*Dummy (Tax Treaty) it +ε it (3) The variable specifications are as follows:
Y: It is the dependent variable to describe Japan’s or Korea’s outward FDI flows to
each country measured in US$; t represents year starting at 1, which means the first year 1985; i represents a country; OFDI is retrieved from OECD statistics.
GDP: It is the year-end gross domestic product in US$; retrieved from UNCTAD. GDP per capita: It is the year-end GDP per capita in US$; retrieved from the World
Bank database.
Distance: It is the straight-line distance between the capitals of the home country and
host country or region in kilometres; retrieved from the Google Map.
Exchange rate: It is a control variable of the annual average exchange rate per US$;
retrieved from UNCTAD.
Inflation rate: It is a control variable of the annual average inflation rate in
percentage; retrieved from the World Bank database.
Dummy tax treaty: It is a dummy variable to decide if the home country has a tax
treaty with a host country it equals 1, other wise 0; retrieved from UNCTAD.
3.3 Sample Description
There are 25 countries or regions I select, and they are divided into two categories: the developed group and the developing group. The third group Asian group is picked out from these 25 countries. Tax havens are excluded in case of bias in the results; the country lists are as follows:
Developed group (13 in total): Australia, Canada, France, Germany, Italy, Japan, South Korea, Netherlands, New Zealand, Spain, Taiwan, UK, US.
Developing group (12 in total): Argentina, Brazil, Chile, China, India, Indonesia, Malaysia, Mexico, Philippines, Russia, South Africa, Thailand.
Asian group (9 in total): China, India, Indonesia, Japan, South Korea, Malaysia, Philippines, Taiwan, Thailand.
The countries I choose are available for the data used for the gravity model (3). The developed, developing, and Asian groups will be applied to the model separately to test if OFDI determinants are different in these groups.
4. Results and Discussions
The following six tables are the regression results of gravity model used in each sub-period from home countries (Japan, South Korea and China) to host countries developing, developed and Asian countries, respectively. The results are supposed to compare with expectations in hypothesis. For all correlation matrixes of the independent variables in each period, see the Appendix 2. The fifteen correlation matrixes express that the independent variables fit well. So no variable factor is eliminated because of the effect of multicollinearity.
Table 4.1.1 Regression Results for OFDI flows to developed countries.
Variable Japan South Korea China
1995-2002 2003-2010 1995-2002 2003-2010 2004-2012 log(GDP) 0.506525*** 0.137634 0.647386*** 0.350282** 0.195525 (3.30924) (1.076193) (3.769107) (1.848634) (0.996719) log(Distance) -2.172839*** -3.491833*** -0.164457 -2.446689*** 0.986339** (-5.143147) (-3.399441) (-0.588473) (-2.804476) (1.693848) Exchange rate -0.000404** -0.000743 -0.000402** -0.020799*** 0.002480*** (-2.247925) (-0.817281) (-2.018503) (-2.600437) (3.751904) Inflation rate -0.016108 0.076461** 0.053100 -0.095582 -0.033823 (-0.331310) (1.755513) (0.846754) (-1.245839) (-0.424207) Tax treaty -0.224511 -0.906178 -0.086522 -0.472334* -0.632917 (-0.70645) (-1.245930) (-0.518816) (-1.340976) (-1.082407)
log(GDP per capit a)
4.933232*** 9.759365*** 2.091611** 6.991671*** 5.951128*** (4.723781) (12.075110) (3.769107) (5.595882) (4.781954)
R2 0.50633 0.748052 0.452181 0.470353 0.496855
N 96 96 96 96 117
Note: *Significance at 10% level; **significance at 5% level; ***significance at 1% level. Each period of Japan or South Korea has 8 years data with 12 developed countries, so the number of observation is 96; China has 9 years data with 13 developed countries, so the number of observation is 117.
Table 4.1.2 Expected and actual results of OFDI flows to developed countries.
Variable Expected Japan South Korea China +/- 1995-2002 2003-2010 1995-2002 2003-2010 2004-2012 log(GDP) + +*** + +*** +** + log(Distance) - -*** -*** - -*** +** Exchange rate + -** - -** -*** +*** Inflation rate - - +** + - -Tax treaty + - - - -*
-log(GDP per capit
a) + +*** +*** +** +*** +***
Table 4.1.1 & 4.1.2 display the regression results for home countries OFDI to developed countries. GDP (log) has a positive impact on all the samples. But China’s distance (log) factor does not comply with the expectation; it shows a positive sign while Japan’s and South Korea’s are negative in each period. Only for China, the impact of exchange rate is in accordance with expectation to be positive, while Japan’s and South Korea’s are on the contrary. For the inflation rate, Japan’s second period and Korea’s first period are positive which is against the expectation. The impact of the tax treaty is negative in all home countries, while the expectation is positive. GDP per capita (log) is consistent with expectations (positive) and is significant. Hereby, China has similarities with Japan and South Korea in factors GDP, inflation rate, tax treaty, and GDP per capita.
Table 4.2.1 Regression Results for OFDI flows to developing countries.
Variable Japan South Korea China
1995-2002 2003-2010 1995-2002 2003-2010 2004-2012 log(GDP) 0.080763 0.550465*** 0.555163*** 0.973119*** 0.512493** (0.407473) (3.691885) (4.114842) (6.323944) (2.207099) log(Distance) -1.121864*** -0.846345*** -0.973425*** -1.481111*** -1.331805*** (-3.512555) (-3.062651) (-4.499185) (-6.129276) (-3.429412) Exchange rate 0.000042 0.000014 0.000055** 0.000078*** 0.000084*** (1.263067) (0.569302) (1.935468) (2.946129) (2.819710) Inflation rate -0.006390** -0.043914*** 0.000312 -0.007064 0.036467* (-2.313253) (-2.410337) (0.138954) (-0.368461) (1.539461) Tax treaty 0.194935 0.335240** 0.094772 0.552151** 0.549176** (1.027869) (1.727759) (0.576213) (1.932776) (1.823136)
log(GDP per cap ita)
R2 0.380245 0.457583 0.573400 0.649163 0.509211
N 96 96 96 96 99
Note: *Significance at 10% level; **significance at 5% level; ***significance at 1% level. Each period of Japan or South Korea has 8 years data with 12 developing countries, so the number of observation is 96; China has 9 years data with 11 developing countries, so the number of observation is 99.
Table 4.2.1 & 4.2.2 display the regression results and expectation from home countries OFDI to developing countries. GDP (log) is the positively impacted in all the samples. Distance (log) factor also complies with the expectation (negative). Exchange rate is also in accordance with expectation to be positive. For the inflation rate, China’s is positive against the expectation, and only South Korea’s in 2003-2010 has the same positive sign with China, but only Japan’s inflation factor manifests the significance. Tax treaty is positive in all home countries, in line with the expectation. GDP per capita (log) is consistent with expectations (negative) in Japan’s two periods and Korea’s first period, but not significant. Korea’s second period and China are positive related and significant, which is the opposite of expectation. Hence, China has similarities with Japan and South Korea in factors GDP, distance, exchange rate, and tax treaty.
Table 4.2.2 Expected and actual results of OFDI flows to developing countries.
Variable Expecte
d Japan South Korea China
+/- 1995-2002 2003-2010 1995-2002 2003-2010 2004-2012 log(GDP) + + +*** +*** +*** +** log(Distance) - -*** -*** -*** -*** -*** Exchange rate + + + +** +** +*** Inflation rate - -** -*** + - + Tax treaty + + +** + +** +**
log(GDP per capit a)
- - - - +** +***
Note: *Significance at 10% level; **significance at 5% level; ***significance at 1% level.
Table 4.3.1 Regression Results for OFDI flows to Asian countries.
Variable Japan South Korea China
1995-2002 2003-2010 1995-2002 2003-2010 2004-2012 log(GDP) -0.383008*** 0.443236*** 0.302815** 0.410934** 0.058339 (-2.49495) (3.092869) (2.151604) (2.309810) (0.288755) log(Distance) -1.271344*** -0.224647 -1.129745*** -1.242776*** -0.197161 (-4.003102) (-0.645569) (-3.568169) (-2.970215) (-0.364975) Exchange rate 0.000003 -0.000030* 0.000039 0.000037* 0.000073*** (0.158399) (-1.611323) (1.140759) (1.489439) (2.487391) Inflation rate 0.005350 0.024035 0.009856 0.022071 -0.000264 (0.835183) (1.063540) (1.032526) (0.694163) (-0.006286) Tax treaty 0.423837*** 0.310109** 0.546555*** 0.766991*** 0.524419* (3.515907) (2.049705) (2.718789) (2.972885) (1.277811)
log(GDP per cap ita) -0.303928*** -0.001667 -0.418754*** -0.281755* 0.290899 (-2.796511) (-0.010817) (-3.059020) (-1.553626) (0.985693) R2 0.275333 0.378477 0.529129 0.528637 0.276566 N 64 64 64 64 72
Table4.3.2 Expected coefficients and real results of Japanese, South Korean, and Chinese OFDI flows to Asian countries.
Variable Expecte d
Japan South Korea China +/- 1995-2002 2003-2010 1995-2002 2003-2010 2004-2012 log(GDP) + -*** +*** +** +** + log(Distance) - -*** - -*** -*** -Exchange rate + + -* + +* +*** Inflation rate - + + + + -Tax treaty + +*** +** +*** +*** +*
log(GDP per capit
a) - -*** - -*** -* +
*Significance at 10% level; **significance at 5% level; ***significance at 1% level.
Table 4.3.1 & 4.3.2 display the regression results and expectation from home countries OFDI to Asian countries. GDP (log) is the positive related except Japan’s first period. Distance (log) factor complies with the expectation; it shows a negative sign to OFDI. Exchange rate is in accordance with expectation to be positive, in spite of Japan’s second period. For the inflation rate, Japan’s and South Korea’s show positive, but China shows a negative relationship which supports the expectation. Tax treaty stays positive and significant in all home countries, and the expectation is also positive. GDP per capita (log) is consistent with expectations (negative) except China (positive). The R-squares of China and Japan are much smaller than South Korea, this means that China’s and Japan’s goodness of fitting is not fairly good. In short, China has similarities with Japan and South Korea in factors GDP, distance, exchange rate, and tax treaty.
4.2 Discussions
model and the flying geese model, and China’s OFDI pattern compared to Japan and South Korea.
4.2.1 Gravity model
As can been seen from the tables 4.1.1- 4.2.3 that Japanese, South Korean and Chinese GDP (log) all have a positive impact on OFDI, both to developed and to developing countries, although it is not always significant. If the sample is restricted to East-Asia, we see the same, but the impact is negative for Japan in the period 1995-2002, which is hard to explain.
Distance between home and host country, is another variable in the core gravity model, which shows a negative impact of Japanese and South Korean OFDI, but not for Chinese OFDI to developed countries. If we look at the sample of the developing and the Asian countries, distance also has a negative impact for Chinese OFDI. So the only exception (positive impact) is China’s OFDI to developed countries, this may be explained as the fact that China’s largest OFDI destination Hong Kong (tax haven) is not included in the developed group; although this sample includes nearby countries Japan and South Korea, they are not China’s main OFDI countries. Instead, some EU countries, US and Australia constitute portions of OFDI, while they are far away from China in geographical location. This may influence China’s results in developed group.
4.2.2 Flying geese model
Distance also plays an important role in the FG-model, because countries with rising wages (like Japan in the 80s, South Korea in the 90s and nowadays China) first transfer labour-intensive activities to cheaper countries nearby, and later further away. This is partly confirmed, because distance has an negative impact for OFDI in Japan, South Korea and China, especially if we restrict the sample to developing and Asian countries. However, the results don't show that as years go by, OFDI moves from countries nearby to countries further away. Maybe distance as such, in miles or kilometers, is not so important because shipping over the ocean is very cheap.
capita instead. Now splitting up between developed and developing countries is crucial.
For developed countries, a high GDP per capita means that people have more money to buy products, so it has a positive impact for OFDI (and export) in countries like Japan, South Korea and China. My results confirm this hypothesis because for all the three countries and for both periods the positive coefficients are significant.
For OFDI to developing countries, GDP per capita is a crude indicator for labour costs. According to the Flying geese model, countries with increasing labour cost will invest in countries with lower labour costs. The results are mixed. In the period 1995-2002 the impact is negative (as expected) but not significant. If we restrict the sample to Asian countries, the impact is highly significant (and negative). For the more recent period (2003-2010) the results are rather different. The impact of GDP per capita is (significant) positive for OFDI to developing countries. Within the Asian sample the results are mixed.
There could be various reasons for these results. One of them may be the Flying geese model only holds for the host countries where wages are lower than in the investing countries (Japan, South Korea and China). If we compare GDP per capita for these three countries with the sample, we get the following result.
Table 4.4 Home country’s average GDP per capita (log) in both periods compared to developing countries’.
1995-2002 2003-2010 2004-2012
Japan 4.527374 4.555133 n.a.
Developing group 3.424502 3.532724 n.a.
South Korea 4.149504 4.294213 n.a.
Developing group 3.423886 3.532724 n.a.
China n.a. n.a. 3.431263
Developing group n.a. n.a. 3.573420
China’s is lower than its developing group average. This means for China that some developing countries may have higher labour costs. While Japan and South Korea have benefited from lower labour costs in other Asian countries nearby (see figure 2.3), China is forced to search for cheaper labours further away, for instance in Africa.
4.2.3 Control variables
Three control variables are used: exchange rate, inflation rate, and tax treaty.
Exchange rate
Exchange rate results reflect that it can influence the OFDI. The exchange rate shows significant and negative relation in developed group, in spite of China’s positive significance. This can be explained as the exchange rate is how much host country’s currency per US$, not host’s per home’s currency. Chinese currency to US$ appreciated since 2006 onwards, but host country’s currency to US$ does not fluctuate so much, resulting in a higher Chinese currency to host countries. This may lead to the positive exchange rate coefficient.
For the developing group, all the exchange rates coefficients are positive, especially South Korea and China display significance. This is consistent with the hypothesis. For Japan, it shows no significance, which means exchange rate is not a dominant factor to influence its OFDI, because Japan already transfers itself into a high-tech motive country, it might has some other driven factors for OFDI to developing countries. In Asia-specific group, exchange rates stay positive except Japan’s period 2003-2010. But the overall coefficients of exchange rate to developed, developing and Asian countries are very small, this tells us that exchange rates do impact OFDI, but impact is very small.
Inflation rate
Inflation rates depict a mix in all these three groups. For developed group, Japan in 2003-2010 and Korea in 1995-2002 do not meet the expectation, while others stay negative but not significant, we can say that inflation rate is not an important factor in OFDI to developed countries.
inflation rates interest OFDI but later on disperse OFDI, but the impact is very limited. China shows a significant and positive relation in OFDI to developing country, this can be due to the higher inflation rate in host countries, China can still get profits, so high inflation still attracts more Chinese OFDI.
For Asian group, inflation rates are positively related regardless of China’s negative results, this phenomenon is hard to explain to my own knowledge. But no any significances exist means inflation rate cannot affect OFDI to Asian countries.
Tax treaty
Most empirical studies confirm the positive role tax treaties play in stimulating FDI, but in the last few years also negative relationships are found. My results show that it is important to make a distinction between developed and developing countries. The OFDI to developed countries is negative influenced by tax treaties, although not significantly. This means that those treaties don't play any role in stimulating FDI, or even a negative role. For developing countries, however, this is different. For Japan, South Korea and China tax treaties do play a positive role for their outward FDI to developing countries, especially in the period from 2003 and 2010. Also in the Asian sample, tax treaties play a significant (positive) role. This may because for developing countries, tax treaties can bring both the traditional benefits (e.g. knowledge and technology spillovers), and increased tax revenue (e.g. business profits and withholding taxes).
4.2.4 China’s outward FDI pattern
in some ways, China follows some experiences of Japan and South Korea, but also create some unique factors to OFDI trend. As Japan and South Korea have higher GDP per capita, it can easily seek Asian countries with cheaper labour costs compared to their own countries, and this really brings them much profit in Asian countries. When it comes to China, as the GDP per capita is on a low level even compared to some Asian countries, the profit China gains in labour-intensive industries is limited, thus China has to move to even further places such as African countries to look for cheap labours with lower wages. Although the transportation costs increase in Africa, due to the cheap cost in ocean transportation, the profit can still be gained.
Figure 4.1 The fraction of China’s outward investment deals by sector, 1998-2011.
Source: Chen and Tang (2014)
Since Chinese government is in a dominant position behind Chinese FDI outflows.
Chen and Tang (2013) also find that Chinese firms are upgrading the skill and capital content of exported products after investment abroad, especially in developed countries. This implies that China follows Japan and South Korea in flying geese pattern that its main OFDI is still in Asia. Figure 4.1 proves that China begins to transfer from labour-intensive (manufacturing) industry to more advanced (e.g. scientific research& technical service) industries, because the portion of each industry is getting larger or smaller as time goes on.
5. Conclusions
The goal of the study is to find empirical evidence of Chinese OFDI pattern from Japanese and South Korea in “Gravity Model” and “Flying Geese Theory” framework.
The study starts with a sample of 25 countries that are divided into developed and developing countries. Then the third sample group Asian countries are selected from the 25 countries to enhance the flying geese pattern within Asian area. The results show that in general, China resembles Japan and South Korea in factors GDP, distance, exchange rate, tax treat, and GDP per capita. But China also has its differences from other two countries, and that can be due to a country’s unique development characteristics. Hypothesis are confirmed except hypothesis 3&4 (exchange rate & inflation rate), because significances are also weak.
inward FDI in labour-intensive industry. Now China has started to expand outward FDI in the world. Some investments are made in Africa in mineral industries, but FDI in labour-intensive manufacturing is also growing.
My findings have important management implications for countries like China, who is experiencing rising labour costs after years of inward FDI. It has been over 20 years since Deng Xiaoping’s well-known Southern Trip in 1992, and China’s “GO Overseas” Policy encourages more Chinese firms to invest abroad. The crucial challenge includes the rising wages and lower-skilled labours. It is very essential for firms to implement structural transformation: from low-skill to high-tech manufacturing, and finally to high-skill services. A number of Chinese firms have shown the feasibility of industry emigration. As more Chinese firms relocate labour-intensive manufacturing in even later “geese”, firms in these geese should perform better in recycling China’s comparative advantage in industries that need higher skills and generate more value-added products. Therefore, Chinese firms should seize the opportunity for industry development, and Chinese government should offer more policy supports, preferential terms, and financial aids to motive firms to invest abroad.
The main limitation of this study is data availability. China and some developing countries have a bad reputation regarding economic and financial data. For that reason, some factors are not included into the gravity model which may lead to some bias. For the flying geese model the labour costs in host countries is a crucial variable but in many countries, these data are not available. In the early years, much data are not available so I have to use the data from 1995. The OFDI stock might be a better indicator than the OFDI flows I was using now. Furthermore, I identify China’s FDI projects myself with limited information so that there might appear mistakes. Also, there are some unregistered Chinese investments to host countries that are excluded in the database. In the future studies, these issues need to be addressed.
and it is also a connection to the world’s trend. From a firm’s level, this topic can help a firm especially a multinational corporation to set better strategic management planning to achieve effective business performance, thus my topic can help with management problems.
As my study is just a start, a great deal of unsolved questions can be studied in the further research. For instance, does the flying geese model apply to European Union, in the sense that labour-intensive activities are transferred from Western to Eastern Europe? If so, what are the differences between East Asia and EU in flying geese model? What else determinants can be applied to the gravity model to help analyse the FDI problem? We encourage future research in this area.
Acknowledgement
Appendix 1
Table A Japan’s OFDI by Industry (12 Asian regions) as Percentage of Share
1951-80 1981-85 1986-90 1986 1987 1988 1989 1990 1951-1990 Food 1.5 1.1 3.3 1.2 2.9 1.6 6.5 1.7 2.5 Textiles 9.4 2.7 2.4 0.8 0.5 2.6 2.2 4.3 3.9 Lumber & Pulp 1.4 0.5 1.2 0.4 0.3 3.2 0.7 1.1 1.1 Chemicals 7.4 6.0 4.7 2.0 5.0 3.6 3.6 7.7 5.5 Iron, nonfer, metals 10.5 7.0 4.0 2.6 6.3 3.7 3.8 3.2 5.9 Machinery 2.8 3.2 3.8 4.1 2.1 4.6 4.3 3.8 3.5 Electric, Electronics 5.5 3.0 12.0 11.3 9.6 15.4 11.4 11.8 8.8 Transport equipment 2.7 4.4 3.6 5.6 4.2 2.8 1.7 5.3 3.6 Others 5.3 2.6 4.7 6.4 3.5 5.1 4.8 4.4 4.4 Manufacturing 46.5 30.6 39.6 34.5 34.5 42.5 39.0 43.2 39.2 Agriculture& Forestry 2.2 0.3 0.3 0.2 0.3 0.3 0.2 0.5 0.7 Fisheries 0.7 0.3 0.3 0.2 0.4 0.4 0.3 0.3 0.4 Mining 30.8 33.3 4.1 10.3 4.9 4.2 2.6 3.3 15.5 Construction 0.8 1.6 1.8 0.6 0.3 1.8 3.5 1.4 1.6 Commerce 4.0 6.8 9.4 8.5 4.0 6.6 8.0 17.4 7.8 Finance &Insurance 2.8 5.5 12.7 12.9 8.2 20.1 13.1 9.2 9.2 Services 6.4 12.7 13.5 25.3 13.8 9.7 13.4 12.4 11.9 Transportation 0.0 2.7 3.1 0.2 3.1 3.4 4.8 1.6 2.3 Real estate 0.0 3.2 9.4 4.1 7.7 6.9 13.6 9.2 6.2 Others 4.6 1.7 3.9 0.4 21.8 0.1 0.1 0.1 3.6 Non-manufacturing 52.3 68.0 58.5 62.7 64.5 53.6 59.8 55.4 59.1 Branch 0.9 1.4 1.9 2.8 1.0 3.9 1.3 1.4 1.6 Acquiring real estate 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Source: Ministry of Finance, Japan
Note: These 12 countries account for more than 99% of Japanese FDI for Asia (Middle East excluded): Brunei, China, Hong Kong, India, Indonesia, Korea, Malaysia, Pakistan, Philippines, Singapore, Taiwan, and Thailand.
Table B Japan’s OFDI by Industry (ASEAN) as Percentage of Share
Mining 48.0 61.7 11.8 42.7 23.2 11.9 7.4 6.4 34.7 Construction 0.7 1.0 2.2 2.4 0.5 1.0 4.4 1.4 1.5 Commerce 1.2 3.8 2.8 7.2 1.5 1.6 3.2 2.8 2.5 Finance &Insurance 1.6 0.7 5.6 7.8 0.7 3.6 9.5 48.0 3.2 Services 1.3 1.7 8.1 2.6 2.3 4.8 8.7 12.3 4.5 Transportation 0.0 0.2 0.8 0.0 0.1 0.8 1.0 0.9 0.4 Real estate 0.0 0.2 5.1 1.0 0.1 3.0 7.9. 6.4 2.4 Others 1.8 0.1 0.1 0.2 0.1 0.0 0.3 2.0 0.6 Non-manufacturing 58.3 70.4 37.7 65.1 30.1 27.7 43.6 36.5 51.8 Branch 0.2 0.2 1.3 0.2 1.6 3.1 0.6 1.0 0.7 Acquiring real estate 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Source: Ministry of Finance, Japan
Table C Japan’s OFDI by Industry (NIEs) as Percentage of Share
1951-80 1981-85 1986-90 1986 1987 1988 1989 1990 1951-1990 Food 1.0 1.0 4.2 0.7 2.8 1.1 9.9 1.3 3.1 Textiles 9.0 0.9 1.2 0.7 0.4 0.8 0.7 3.5 2.4 Lumber & Pulp 0.7 0.1 0.3 0.1 0.1 0.3 0.2 0.7 0.3 Chemicals 10.8 11.5 3.7 2.1 7.4 3.1 2.8 3.6 6.2 Iron, nonfer, metals 3.7 1.5 2.2 2.5 2.5 1.8 1.9 2.4 2.3 Machinery 6.5 6.0 2.3 3.5 3.3 1.8 1.9 2.4 2.3 Electric, Electronics 12.2 4.8 7.8 13.2 10.8 8.3 5.3 602 7.9 Transport equipment 4.0 3.5 2.5 6.6 3.4 2.2 1.6 1.4 2.9 Others 8.5 3.0 39.0 7.9 3.3 4.5 3.1 3.0 4.4 Manufacturing 56.4 32.4 28.0 37.4 34.0 23.7 27.5 24.0 33.1 Agriculture& Forestry 1.2 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.2 Fisheries 0.0 0.1 0.0 0.0 0.0 0.0 0.1 0.0 0.0 Mining 0.2 0.1 0.0 0.0 0.0 0.0 0.0 0.1 0.1 Construction 0.9 2.2 1.9 0.0 0.4 2.5 3.3 1.3 1.8 Commerce 9.2 10.7 14.9 9.0 6.7 10.0 11.5 33.5 13.3 Finance &Insurance 4.8 11.8 19.1 16.8 15.1 32.1 16.4 144.0 15.6 Services 15.4 23.1 15.6 30.8 22.0 11.7 14.0 99.0 16.9 Transportation 0.1 5.7 4.9 0.3 5.8 5.3 7.2 24.0 4.3 Real estate 0.0 7.0 13.0 5.0 12.1 9.9 18.2 12.7 10.0 Others 9.3 3.9 0.0 0.0 2.6 0.2 0.0 0.0 2.4 Non-manufacturing 41.2 64.6 69.9 61.8 64.8 71.7 70.8 74.6 64.6 Branch 2.0 3.0 2.1 0.8 1.2 4.6 1.7 1.5 2.2 Acquiring real estate 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Appendix 2
Correlation Matrix for Model (3): The following 15 tables show correlation matrixes of all the independent variables in each period, and *denotes variable in natural logarithm term. The 15 correlation matrixes express that the independent variables fit well. So no variable factor population is eliminated in order to disregard the effect of multicollinearity, and to verify more accurate analysis.
Table 1 Japan to developed group in 1995-2002
GDP* GDP per
capita* Distance* Exchange Inflation Tax treaty
GDP* 1 GDP Per capita* 0.4575 1 Distance* 0.3267 0.4572 1 Exchange -0.0114 -0.3854 -0.3760 1 Inflation -0.0526 -0.1827 -0.1937 0.3894 1 Tax treaty 0.3158 0.4455 0.4148 -0.0676 -0.0019 1
Table 2 Japan to developed group in 2003-2010
GDP* GDP per
capita* Distance* Exchange Inflation Tax treaty
GDP* 1 GDP per capita* 0.4474 1 Distance* 0.3120 0.3972 1 Exchange -0.0891 -0.3999 -0.3183 1 Inflation -0.0073 -0.0035 -0.0973 0.2751 1 Tax treaty 0.3309 0.4594 0.4022 0.0629 0.2434 1
Table 3 South Korea to developed group in 1995-2002
Table 4 South Korea to developed group in 2003-2010
GDP* GDP per
capita* Distance* Exchange Inflation Tax treaty
GDP* GDP per
capita* Distance* Exchange Inflation Tax treaty
GDP* 1 GDP per capita* 0.3845 1 Distance* -0.0119 0.4324 1 Exchange 0.2126 -0.0992 -0.3884 1 Inflation -0.1538 0.0206041 0.4023 -0.4407 1 Tax treaty 0.3019 0.6497742 0.4397 -0.1194 0.0389 1
Table 5 China to developed group in 2004-2012
GDP* GDP per capita*
Distance* Exchange Inflation Tax treaty
GDP* 1 GDP per capita* 0.3623 1 Distance* -0.0060 0.4049 1 Exchange -0.0929 -0.4506 -0.3921 1 Inflation -0.2324 -0.1918 0.1794 0.1907 1 Tax treaty 0.1928 0.0823 0.3070 0.0362 0.0929 1
Table 6 Japan to developing group in 1995-2002
GDP* GDP per capita*
Distance* Exchange Inflation Tax treaty
GDP* 1 GDP per capita* -0.1590 1 Distance* -0.0740 0.4396 1 Exchange -0.0695 -0.2649 -0.0836 1 Inflation 0.1860 0.0777 0.0753 0.0827 1 Tax treaty 0.4292 -0.0480 -0.0298 0.0916 0.1008 1
Table 7 Japan to developing group in 2003-2010
GDP* GDP per
capita* Distance* Exchange Inflation Tax treaty
GDP* 1 GDP per capita* -0.1198 1 Distance* -0.1704 0.3579 1 Exchange -0.0853 -0.3459 -0.0870 1 Inflation 0.1401 -0.0086 0.2476 0.1776 1 Tax treaty 0.4519 -0.1163 -0.1540 0.0716 0.2203 1
Table 8 South Korea to developing group in 1995-2002
capita* GDP* 1 GDP per capita* -0.1128 1 Distance* -0.1080 0.4454 1 Exchange -0.0750 -0.2465 -0.0486 1 Inflation 0.1786 0.0770 0.0816 0.1068 1 Tax treaty 0.2405 -0.4073 -0.4069 0.0806 0.0788 1
Table 9 South Korea to developing group in 2003-2010
GDP* GDP per
capita* Distance* Exchange Inflation Tax treaty
GDP* 1 GDP per capita* -0.0971 1 Distance* -0.2485 0.4332 1 Exchange -0.0897 -0.3498 -0.0480 1 Inflation 0.1393 -0.0054 0.2532 0.1582 1 Tax treaty 0.1103 -0.1840 -0.3386 0.0865 -0.3757 1
Table 10 China to developing group in 2004-2012
GDP* GDP per
capita* Distance* Exchange Inflation Tax treaty
GDP* 1 GDP per capita* 0.0170 1 Distance* 0.1549 0.4013 1 Exchange -0.0240 -0.3916 -0.1425 1 Inflation 0.3114 -0.0909 0.0974 0.1204 1 Tax treaty 0.2995 -0.3653 -0.4418 0.0624 0.2524 1
Table 11 Japan to Asian group in 1995-2002
GDP* GDP per
capita* Distance* Exchange Inflation Tax treaty
Table 12 Japan to Asian group in 2003-2010
GDP* GDP per capita*
Distance* Exchange Inflation Tax treaty
GDP* 1 GDP per capita* 0.0558 1 Distance* -0.4294 -0.3681 1 Exchange -0.0706 -0.2291 0.3058 1 Inflation -0.0139 -0.4082 0.4403 0.3800 1 Tax treaty 0.2646 -0.2695 0.2587 0.1955 0.2534 1
Table 13 South Korea to Asian group in 1995-2002
Table 14 South Korea to Asian group in 2003-2010
GDP* GDP per capita*
Distance* Exchange Inflation Tax treaty
GDP* 1 GDP per capita* 0.3675 1 Distance* -0.3638 -0.4434 1 Exchange -0.1380 -0.2836 0.4228 1 Inflation -0.2815 -0.3606 0.3879 0.3875 1 Tax treaty 0.0846 -0.4361 0.3028 0.1355 0.2465 1
Table 15 China to Asian group in 2004-2012
GDP* GDP per capita*
Distance* Exchange Inflation Tax treaty
GDP* 1 GDP per capita* 0.3940 1 Distance* -0.3440 -0.4056 1 Exchange -0.0890 -0.2986 0.4030 1 Inflation -0.3189 -0.4176 0.3477 0.4386 1 Tax treaty 0.0093 -0.3093 0.2165 0.1024 0.1849 1 GDP* GDP per
capita* Distance* Exchange Inflation Tax treaty
References
Akamatsu, K. (1935). Waga kuni yomo kogyohin no susei. Shogyo Keizai Ronso, 13, 129-212.
Akamatsu, K. (1962). Historical pattern of economic growth in developing countries.
The Developing Economies, 1, 3-25.
Anderson, J. E. (2010). The gravity model. National Bureau of Economic Research,
16576.
Van Bergeijk, P. A. G. & Brakman, S. (2010). The gravity model in international trade: advances and applications. Cambridge: Cambridge University Press.
Baker, P. L. (2012). An analysis of double taxation treaties and their effect on foreign direct investment. Faculty of Economics, University of Cambridge.
Bano, S. S., Takahashi, Y., Scrimgeour, F. (2013). ASEAN-New Zealand trade relations and trade potential: Evidence and analysis. Journal of economic
integration, 28(1).
Bernard, M. & Ravenhill, J. (1995). Beyond product cycles and flying geese: regionalization, hierarchy, and the industrialization of East Asia. World Politic, 47, 2.
Buckley, P.J., Clegg, L.J., Adam, R.C., Liu, X., Voss, H. & Zhang, P. (2007). The determinants of Chinese outward foreign direct investment, Journal of
International Business Studies, 38(4), 499-518.
Chen, W. & Tang, H. (2013). Export promotion of ODI from emerging markets – transaction-level evidence from China. Johns Hopkins University memo.
Chen, W. & Tang, H. (2014). The dragon soars: micro evidence on Chinese outward direct investment. VoxEU Column.
Cheng, L. & Ma, Z. (2010) China’s outward foreign direct investment.China’s
Growing Role in World Trades, 14, 545-578.
Christians, A. (2005). Tax treaties for investment and aid to Sub-Saharan Africa, a case study. The Berkeley Electronic Press, 154.
Coppel, J. (2008). China’s outward foreign direct investment. Results of the Work Of
the OECD Investment Committee, 6.
neoclassical world? In The Regionalization of the World Economy. Chicago:
University of Chicago Press.
Dowling, M. & Cheang, C. T. (2000). Shifting comparative advantage in Asia: new tests of the “flying geese” model. Journal of Asian Economics, 11, 443-463. FDI in Figures. (2014). International investment struggles: Q3 FDI flows are up, but
2013 is heading towards a second annual decline. OECD.
Froot, K.A. & Stein, J.C. (1991). Exchange rates and foreign direct investment: An imperfect capital markets approach. The Quarterly Journal of Economics 106(4), 1191-1217.
Geda, A. & Meskel, A. G. (2007). China and India growth surge: is it a curse of blessing for Africa? The Case of Manufacturing Exports, Open University.
Di Giovanni, J. (2005). What drives capital flows? The case of cross-boarder M&A activity and financial deepening. Journal of International Economics, 65(1),
127-149.
Jun, K.W., Sader, F., Horaguchi, H. & Kwak, H. (1993). Japanese foreign direct investment: recent trends, determinants and prospects. Debt and International
Finance Division, The World Bank.
Kim, J.M. & Rhe, D.K. (2009). Trends and determinants of South Korean outward foreign direct investment. The Copenhagen Journal of Asian Studies,
27(1),126-154.
Kim, N.T. & Hirata, J. (2009). Foreign direct investment and the export-oriented industrialization in Vietnam. The Ritsumeikan Economic Review, 57(5-6),
574-59.5
Hattari, R. & Rajan, R.S. (2008). Sources of FDI flows to developing Asia: The role of distance and time zones. ADBI Working Paper 117. Tokyo: Asian Development
Bank Institute.
Hobday, M. (1995). Innovation in East Asia: The Challenge to Japan. Edward Elgar
Published Limited.
Human Development Index. (2014). Human Development Report. United Nations. Retrieved from: http://hdr.undp.org/en/content/human-development-index-hdi Isard, W. (1954). Location theory and trade theory: short-run analysis. The Quarterly
Journal of Economics, 68, 2, 205-320.
foreign direct investment. The Copenhagen Journal of Asian Studies, 27,1.
Kojima, K. (1960). Capital accumulation and the course of industrialisation, with special reference to Japan. The Economic Journal, LXX, 757-768.
Kojima, K. (2000). The “flying geese” model of Asian economic development: origin, theoretical extensions, and regional policy implications. Journal of Asian
Economics, 11, 375–401.
Kumagai, S. (2008). A journey through the secret history of the flying geese model.
Academic Research Repository at the Institute of Developing Economies, No. 158.
Linnemann, H. (1966). An econometric study of international trade flows.
Contributions to Economic Analysis, 42.
Ministry of Finance. (2014) Japan’s OFDI data retrieved from http://www.mof.go.jp/english/international_policy/reference/itn_transactions_in_s ecurities/fdi/index.htm
Morck, R., Yeung, B., & Zhao, M. (2007). Perspectives on China's outward foreign direct investment. Journal of International Business Studies, 39, 337-350.
Petri, P. A. (1988). Korea's export niche: origins and prospects. World Development,
16(1), 47-63
Petri, P. A. (2010). The determinants of bilateral FDI: Is Asia different? International
Conference on Investment Technology Spillovers and East Asian FTA Proceeding.
Qu, Y., Cai, F., Zhang, X. (2013). Has the “flying geese” paradigm occurred in China?
China Economist ,8,6.
Ravenstein, E.G. (1885). The laws of migration. J. Stat. Soc. 48, 167–227.
Tinbergen, J. (1962). Shaping the world economy; Suggestions for an international economic policy. Twentieth Century Fund, New York.
World Economic Outlook. (2014). Recovery strengthens, remain uneven.
