F1. Please consider this paper as a competitive one.
F2. For correspondence please use the following address: P. Pantelidis, 39 Samou str. GR-15125, Marousi, Greece. Tel: 30-1-6824692, fax: 30-1-4179064.
The aim of this paper is to test the hypothesis that the outward FDI position of countries may be considered as a function of country specific characteristics, such as technology, supply and demand conditions, etc. The model developed identifies the main determinants of outward FDI using time series data for three European Union members and three developing countries. The model indicates that real GNP is proved the most important determinant of outward FDI. Exchange rate is also one of the main factors influencing outward FDI. Developed European countries specialise in human capital intensive FDI, while developing countries in technology intensive one. Germany and the Republic of Korea seems to adopt a model of internationalisation with FDI following exports. Overall, the results verify that the outward FDI position of countries is influenced by national characteristics and that the same type of endowments have different significance for different countries. The results do not reveal systematic differences between developed and developing countries.
1. INTRODUCTION
Recent contributions to economic theory explaining the outward Foreign Direct Investment (FDI) position of countries suggest that the mix of ownership (O), location (L), internalisation (I) advantages1 of a country's firms differentiates along the country's course of economic development2. If it is accepted that the propensity of a country's firms to invest abroad is a function of their ability to acquire and utilise internally income yielding assets, in the sense that the higher this ability is the higher the degree of foreign involvement will be, and if in turn this ability of firms is a function of assets, either natural, e.g. natural resources, unskilled labour, etc. or created, e.g. capital, technology, skilled labour, then the propensity of investing abroad can be hypothesised to be a function of such endowments, which are country specific, in the sense that different countries possess different natural endowments and create different technological and human capital assets through the adoption of different policies and the following of idiosyncratic development courses. In the line of the above argument country specific assets are dynamic in character, evolving along the course of the country's development, as the result of the interplay between policies, past levels of developments and actions of economic agents. Firms may draw upon the supply of these endowments in order to organise production efficiently, create their own advantages and be able to supply domestic and foreign markets. As the configuration of country specific endowments change over time the firm specific ability to serve markets also changes and so it does the propensity to invest abroad.
The aim of this paper is to test the hypothesis that the outward FDI position of countries may be considered as a function of country specific characteristics, such as technology, supply and demand conditions, etc. The model developed identifies the main determinants of outward FDI using time series data for six countries. Three European Union (EU) members, namely France, Germany and Italy and three developing countries: Brazil, the Republic of Korea and Singapore. The period of investigation is between 1976 and 1997.
Table 1 shows that France, Germany and Italy accounted for the 43.3% of total EU outward FDI stock in 1995, the 40.9% in 1990 and 34.6% in 19803. The increase is due to the growth of the individual shares of France and Italy between 1980 and 1995 from 11% to 16.6% and from 3.4% to 7.2% respectively, whereas Germany maintained a rather stable share around 20.0% in the same period4. Brazil, the Republic of Korea and Singapore although they maintained a share of around 15.0% of all developing countries outward FDI stock between 1980 and 1995 changed their individual positions considerably. Brazil lost a significant part of its share moving from 10.6% in 1980 to only 3.0% in 1995, while Singapore and Korea increased their own from 2.8% and 2.3% to 6.5% and 5.2% respectively in the same period5.
Table1: Outward FDI stock by home country (in millions of $).
Home country | 1980 | 1985 | 1990 | 1995** |
France | 23604 (11.0%) | 37077 (12.9%) | 110126 (14.2%) | 200902 (16.6%) |
Germany | 43127 (20.2%) | 59909 (20.9%) | 151581 (19.5%) | 235003 (19.5%) |
Italy | 7317 (3.4%) | 16301 (5.7%) | 56102 (7.2%) |
86672 (7.2%) |
European Union* | 213157 (100%) | 286485 (100%) | 777227 (100%) | 1208838 (100%) |
Korea |
142 (2.3%) |
526 (2.5%) |
2095 (3.0%) |
11079 (5.2%) |
Brazil | 652 (10.6%) | 1361 (6.4%) | 2397 (3.5%) |
6460 (3.0%) |
Singapore |
173 (2.8%) |
1320 |
4741 (6.8%) |
13842 (6.5%) |
Developing countries | 6167 (100%) | 21222 (100%) | 69369 (100%) |
214453 (100%) |
Source: UNCTAD (1996), World Investment Report, Annex table II, pp. 245-248.
Dependent Variable
Annual FDI outflows.
Independent Variables
Income
As the income of a country rises its economic structure changes and so it does the mix of the country’s competitive advantages. A growing share of the GNP is accounted for by manufacturing and services, the capital intensity of production increases, demand patterns move towards the consumption of differentiated products and markets grow. The latter improves the realisation of economies of scale through specialisation, the introduction of new technology and greater volumes of output6.
Firms taking advantages of the country specific agglomeration advantages develop their ownership advantages. For example, sophisticated demand patterns motivate firms, especially in consumer goods and services to differentiate products and improve their marketing expertise. That in turn may be a significant competitive advantage in establishing foreign production especially in markets with demand conditions requiring local product adaptation7. As firms accumulate ownership specific advantages their propensity to undertaking direct foreign production increases, especially if these advantages are intangible and thus better transferred abroad through the creation of an internal market rather than via an arm's length type of transaction (internalisation advantages)8.
The expectation is that higher income levels of a country are associated with greater outward FDI activity. Real GNP is proposed as the proxy for a country's level of income and structural transformation.
Interest Rate
Foreign operations require significant commitment in capital, especially if they are undertaken in capital intensive sectors where production is characterised by extensive economies of scale, as the case is for most of FDI. The capital abundance of the home country may form the necessary background for establishing large firms with adequate financial means and relatively easy access to capital markets. Capital abundance is associated with relatively low interest rates, which in turn decreases the opportunity cost of capital. That may prove investments abroad profitable although the risks and uncertainties associated with such investments. The hypothesis is that the lower the interest rate of the home country the higher the country's propensity for outward FDI is9.
Exchange Rate
Aliber (1970) argued that firms from countries with strong currencies are able to support financially their foreign investments in better terms than firms from countries with weak currencies. The appreciation of the home country currency lowers the capital requirements of foreign investments in domestic currency units enabling the investing abroad firm to raise capital easier than in the case of a depreciated currency. Besides, the home currency appreciation reduces the nominal competitiveness of exports, increasing that way the motive for choosing FDI as the mode of servicing foreign markets. A negative correlation between exchange rate and outward FDI is hypothesised and the effective exchange rate of the home country is proposed as an approximation of the variable.
Technology
The hypothesis that technological capability is positively related with FDI has received extensive theoretical and empirical support10. The ability to organise and undertake the production of technological inputs is a critical ownership competitive advantage yielding income for the possessing firm. Markets fail to optimise the returns on technological inputs transactions, especially if technology is information intensive11. In that case the exploitation of technological intermediate goods across national boundaries is internalised by firms via FDI.
The ability of firms to organise and produce technological inputs varies across countries according to characteristics such as the legal and patent systems, availability of inputs and skills necessary for the production of technology, market structure, government policies and incentives in education, scientific research, etc. The number of patents issued in a country is expected to approximate the ability of firms to generate technological inputs and thus it is positively related with the outward FDI propensity of the country.
Human Capital
Human skills is another powerful ownership advantage the possession of which gives the ability for acquiring competitive advantages of other type. R&D, marketing, management and organisation activities and foreign operations require competent and skilled labour. Empirical research has proved that FDI is more likely in skill intensive sectors12. Again, the human capital supply varies across countries according to education and training systems, government policies, etc.
The approximation proposed for this variable is higher education R&D personnel. Because, this data is not available for the three developing countries the third level education students is suggested as a substitute. The higher the third level education graduates the higher the skill content in employment is expected to be, and so it is suggested a positive relation between the variable and FDI.
The Openness of the Economy
The liberalisation of a country's foreign economic transactions is expected to influence positively the outward FDI activities of its firms. First, the absence of capital controls allows the unrestricted funding of investments abroad13. Second, an export oriented economy permits firms to acquire information about foreign markets, e.g. demand and supply conditions, the legal system, etc., knowledge and skills about organising foreign operations and marketing their products internationally. All these may form the background for changing the mode of internationalisation from exporting to FDI14. Third, firms may choose to combat import competition via increasing their involvement in the home markets of the import producing companies and a certain mode of retaliation is FDI. Overall, an higher degree of openness is expected to be associated with a higher level of outward FDI activity.
The exports plus imports level of a country is proposed as an approximation of its openness.
Dummy Variable
The unification of Germany is expected to act negatively on the level of German outward FDI shifting investments towards Eastern Germany for modernising its economy. The dummy variable takes the value of zero for the years up to 1990 and the value of one for the years afterwards.
3. METHODOLOGY AND DATA
The model function can be summarised as it follows:
FDI = F | ( | Y, | I, | ER, | TE, | HC, | OP, | D | ) | |
+ | - | + | + | + | + | - |
where:
FDI = outward flows of FDI
Y = home country real GNP
I = home country interest rate
ER = home country effective exchange rate
TE = technology variable. It is approximated by the number of patents issued in the home country.
HC = Human capital variable. It is approximated by the number of higher education R&D personnel for the three EU countries and by the number of third level education students for the three developing countries.
OP = Openness of the economy. It is approximated by the level of exports plus imports
D = dummy variable applicable in the case of Germany and measuring the impact of German unification: It takes the value 0 for the years between 1976 and 1990 and the value 1 for the years between 1991 and 1997.
The signs underneath the variables indicate the expected type of correlation (negative or positive) between the independent variables and FDI outflows. The loglinear form of the equation is estimated using OLS for each country separately with annual data for the period 1976-1997. The equation has a loglinear form because under this specification elasticities given by the estimated coefficients are constant.
The FDI, real GNP, exchange rate, interest rate, export and import variables have been taken from IMF International Financial Statistics. The source of the technology and human capital variables in the case of the EU countries is the OECD Main Science and Technology Indicators and in the case of developing countries is the United Nations Statistical Yearbook.
The stationarity of all used data series has been tested by applying the Plillips-Perron unit root test. It was found that all European countries and the Republic of Korea series are of level zero, while Brazil and Singapore series are of level one15. The result of all unit root test are reported in Table 3. There is also no strong indication of multicollinearity, since all the statistically significant coefficients have the expected signs. The Durbin-Watson statistic indicates the absence of autocorrelation16.
4. RESULTS
The results are presented in Table 2. The constant term is significant in all cases, with the exception of Singapore indicating that a part of outward FDI is due to strategic considerations of firms in an increasingly globalised business environment.
In the case of France the statistically significant variables and with the expected signs are: income, technology, at the 10% level, and human capital. Exchange rate is significant, but its correlation with FDI outflows is positive. If the devaluation of the French Franc is made in order to adjust for losses in competitiveness in terms of productivity or labour cost or inflation, then French firms may take the option of transferring abroad the less sophisticated and more vulnerable of increases in labour costs parts of their production processes, thus securing their shares in both the domestic and exporting markets through rationalised FDI where strategic rather than capital considerations are taken into account17.
German outward FDI is best explained by: income, exchange rate, at the 10% level, human capital and the openness of the economy. These variables also have the expected sign. The dummy variable is significant at the 10% level and has the expected negative sign explaining perhaps the stability of the German share in total European Union FDI stock in recent years (see Table 1). The shift of investments towards the Eastern part of the country lowered the rate of increase of FDI outflows relatively to other European countries.
Income, interest rate, at the 10% level, and human capital are the statistically significant and with the expected signs determinants for the Italian outward FDI .
Interest rate and openness are the only significant and with the expected signs explanatory variables in the case of Korea, while only one variable, i.e. technology being the significant and positively correlated with outward FDI determinant for both Brazil and Singapore.
The exchange rate is statistically significant at the 10% level with a positive sign for both Brazil and Singapore. That mean that if domestic currency depreciates, outward FDI increases. If currency depreciation compensates for deteriorating productivity and labour cost export oriented FDI may be used as a long term effective measure for securing market shares abroad. This type of FDI is directed towards countries of more favourable cost structures, mainly other developing countries. Indeed, locations in developing countries accounted for almost the 46.0% of outward Brazilian FDI stock in 1993 and the same applies to Singapore, whose FDI locations concentrate to a significant degree in other Asian, particularly ASEAN countries and China18. In the case of Singapore perhaps 50.0% of its outward FDI is undertaken by foreign subsidiaries located there19. It is plausible to assume that if their main motive for establishing in Singapore is exporting, then any increase in labour cost drives them to invest in other countries in the region in order to take advantage of cost differentials and continuing that way their exporting activity.
Income is significant at the 10% level and has the expected positive relation with FDI outflows in the case of Brazil.
Table 2: OLS estimates of outward FDI for period 1976-1997.
France |
Germany |
Italy |
Korea |
Brazil |
Singapore |
|
constant |
-83.0* |
-27.2* |
-52.2* (2.91) |
-14.7** |
-48.5** |
-14.1 |
Y |
10.2* |
9.2* |
11.5* |
-0.3 |
5.2 ** |
-8.8 |
I |
0.2 |
0.2 |
-0.5** |
-1.6* |
0.05 |
-0.5 |
ER |
3.6* |
-0.1** |
1.0 |
0.8 |
0.1** |
8.4** |
TE |
3.9** |
-0.004 |
0.05 |
-0.3 |
1.08* |
3.2* |
HC |
5.5* |
4.11* |
2.7** |
0.01 |
-0.3 |
4.6 |
OP |
-0.5 |
0.6** |
0.5(0.5) |
2.3* |
0.3 |
0.6 |
D |
-0.45** (1.82) |
|||||
R2 |
0.96 |
0.95 |
0.92 |
0.94 |
0.63 |
0.88 |
F stat. |
63.4 |
51.1 |
23.8 |
46.1 |
3.8 |
18.5 |
DW |
2.07 |
2.04 |
2.05 |
1.89 |
2.12 |
2.31 |
Table 3: Phillips-Perron unit root test (t-stat.)
France |
Germany |
Italy |
Korea |
Brazil |
Singapore |
|
FDI |
-4.81* |
-4.68* |
-11.55* |
-5.94* |
-4.98* |
-9.30* |
Y |
-3.13* |
-2.68** |
-2.66** |
-8.64* |
-3.64* |
-2.88** |
I |
-3.17* |
-3.30* |
-2.66** |
-3.12* |
4.04* |
-3.05* |
ER |
-2.69** |
-5.10* |
-3.83* |
-2.67** |
-6.91* |
-3.05* |
TE |
-3.94* |
-3.91* |
-2.74** |
-3.93* |
-4.22* |
-3.30* |
HC |
-3.26* |
-2.65** |
-3.15* |
-2.73** |
-20.31* |
-3.34* |
OP |
-6.65* |
-3.65* |
-2.91** |
-2.94** |
-2.67** |
-2.70** |
D |
-4.35* |
5. CONCLUSIONS
The model has a good explanatory ability relatively better in the case of the developed European countries than the developing ones.
Real GNP is proved the most important determinant of outward FDI. Exchange rate is also one of the main factors influencing outward FDI. It seems though that it is rather indirectly connected with FDI revealing changes of national competitiveness, with the exception of Germany.
Developed European countries specialise in human capital intensive FDI, while developing countries in technology intensive one. Germany and the Republic of Korea seems to adopt a model of internationalisation with FDI following exports. In both cases firms use FDI for competing import penetration either supplying their home markets with cheaper production abroad or threatening the position of competitors in their domestic markets20.
Overall, the results verify that the outward FDI position of countries is influenced by national characteristics and that the same type of endowments have different significance for different countries. The results do not reveal systematic differences between developed and developing countries.
NOTES
1. See Dunning (1993), pp. 76-86.
2. See Dunning and Narula (1996).
3. Calculations are made on the basis of the following 13 EU countries for all years for reasons of comparability: Austria, Belgium and Luxembourg, Denmark, Finland, France, Germany, Ireland, Italy, Netherlands, Portugal, Spain, Sweden, United Kingdom.
4. See Table 1.
5. It should be noted that the Singapore's share was stable at around 6.5% between 1985 and 1995. All three developing countries increased the absolute amount of their outward FDI stock with the Republic of Korea making the highest performance. Its stock was increased more than five tines between 1990 and 1995 and four times between 1985 and 1990, while Singapore achieved an increase of three and 3.6 times respectively and Brazil only 2.7 and 1.7 times respectively. See Table 1.
6. See Chenery et al. (1986).
7. See indicatively Caves (1971), Lall (1980), Grubaugh (1987).
8. See the eclectic paradigm, Dunning (1993), pp. 76-86.
9. A number of researchers have conducted tests on the positive and statistically significant association between capital requirements or capital intensity and FDI with uneven results as between countries for the same study, see for instance Clegg (1987) and for the same country between studies, see for instance Prugel (1981) and Lall (1980), Grubaugh (1987).
10. See indicatively Lall (1980), Pugel (1981), Grubaugh(1987), Clegg (1987), Cantwell (1981, 1987), Pearce (1989), Kogut and Chang (1991), Dunning (1993), etc.
11. See for instance Buckley and Casson (1976, 1985), Dunning (1993).
12. See indicatively Juhl (1979), Lall (1980), Pugel (1981), Clegg (1987), etc.
13. There is some empirical research on the effectiveness of the US capital control programme implemented between 1965 and 1974. See Scaperlanda and Mauer (1973), Scaperlanda and Balough (1983), Scaperlanda (1992).
14. See Kogut (1983) for an analysis on the sequential internationalisation of firms.
REFERENCES
Aliber, R. Z., 1970, A Theory of Foreign Direct Investment. In C. P. Kindleberger (ed.), The International Corporation. Cambridge, MA: MIT Press.
Buckley, P. J. and Casson, M., 1976, The Future of the Multinational Enterprise. London: Macmillan.
Buckley, P. J. and Casson, M., 1985, The Economic Theory of the Multinational Enterprise. London: Macmillan.
Cantwell, J., 1981, Technological Advantage as a Determinant of the International Economic Activity of Firms, Discussion Papers in International Investment and Business Studies, No 105. Reading: University of Reading.
Cantwell, J., 1987, The Reorganisation of European Industries after Integration: Selected Evidence on the Role of Transanational Enterprise Activities, Journal of Common Market Studies, 36, pp. 127-152.
Caves, R. E., 1971, Industrial Corporations: the Industrial Economics of Foreign Investment, Economica, 38, pp. 1-27.
Chenery, H.B., Robinson, S. and Syrquim, M., 1986, Industrialisation and Growth: Comparative Study. New York: Oxford University Press.
Clark, D. P., Sawyer, W. C. and Sprinkle, R. L., 1989, Determinants of Industry Participation under Offshore Assembly provisions in the US Tariff Code, Journal of World Trade, 23(5), October, pp. 123-130.
Clegg, J., 1987, Multinational Enterprises and World Competition. London: Macmillan.
Dunning, J. H., 1993, Multinational Enterprises and the Global Economy. Workingham Berks: Addison-Wesley.
Dunning, J.H. and Narula, R., 1996, The Investment Development Path Revisited: Some Emerging Issues. In J. Dunning and R. Narula (eds.), Foreign direct Investment and Government. Catalysts for Economic Restructuring. London: Routledge.
Dunning, J. H., Van Hoesel, R. and Narula, R., 1996, Explaining the «New» Wave of Outward FDI for Developing Countries: The Case of Taiwan and Korea. In Innovation and International Business, 22nd Annual EIBA Conference Procceedings. Stockholm: Institute of International Business.
Grubaugh, S. J., 1987, Determinants of Direct Foreign Investment, Review of Economics and Statistics, 69(1), pp. 149-152.
Juhl, P., 1979, On the Sectoral Patterns of West German Manufacturing Investment in Less Developed Countries: the Impact of Firm Size, Factor Intensities and Protection, Weltwirtschaftliches Archiv, 115(3).
Kogut, B., 1983, Foreign Direct Investment as a Sequential Process. In C. P. Kindleberger and D. P. Audretsch (eds.), The Multinational Corporations in the 1980s. Cambridge, MA: MIT Press.
Kogut, B. and Chang, S. J., 1991, Technological Capabilities and Japanese Direct Investment in the United States, Review of Economics and Statistics, LXXIII, pp. 401-413.
Kojima, K., 1978, Direct Foreign Investment: a Japanese Model of Multinational Business Operations. London: Croom Helm.
Kojima, K. and Ozawa, T., 1985, Towards a Theory of Industrial Restructuring and Dynamic Comparative Advantage, Hitotsubashi Journal of Economics, 26(2).
Lall, S., 1980, Monopolistic Advantages and Foreign Involvement by US Manufacturing Industry, Oxford Economic Papers, 32, pp. 102-122.
Moxxon, R., 1975, The Motivation for Investment in Offshore Plants: the Case of the US Electronics Industry, Journal of International Business Studies, 6(1), pp. 51-65.
Pearce, R. D., 1989, The Internationalisation of Sales by Leading Enterprises: Some Firm, Industry and Country Determinants, Discussion Papers in International Investment and Business Studies, Series B, No 135. Reading: University of Reading.
Phillips, P.C.B. and Person, P., 1988, Testing for a Unit Root, Biometrica, 75, pp. 335-346.
Pugel, T. A., 1981, The Determinants of Foreign Direct Investment: an Analysis of US Manufacturing Industries, Managerial and Decision Economics, 2, pp. 220-228.
Scaperlanda, A., 1992, Direct Investment Controls and International Equilibrium: the US Experience, Eastern Economic Journal, 18.
Scaperlanda, A. and Balough R., 1983, Determinants of US Direct Investment in the EEC Revisited, European Economic Review, 21, pp. 381-390.
Scaperlanda, A. and Mauer, L. J., 1973, The Impact of Controls on US Direct Foreign Investment in the EEC, Southern Economic Journal, 39, pp. 419-423.
UNCTAD, 1995, World Investment Report. New York: United Nations.
UNCTAD, 1996, World Investment Report. New York: United Nations.