the investment development path and the product cycle: an ... · building on vernon’s product...

gareth.jones Section name

The investment development path and the product cycle: an integrated approach

The investment development path and the product cycle: an integrated approach

by by Constantina Kottaridi, Fragkiskos Filippaios and Marina Papanastassiou Constantina Kottaridi, Fragkiskos Filippaios and Marina Papanastassiou

2004 003 Henley Business School University of Reading Whiteknights Reading RG6 6AA United Kingdom www.henley.reading.ac.uk

The investment development path and the product cycle - an integrated approach: Empirical evidence from the new EU member states of CEE

1

THE INVESTMENT DEVELOPMENT PATH AND THE

PRODUCT CYCLE - AN INTEGRATED APPROACH:

EMPIRICAL EVIDENCE FROM THE NEW EU MEMBER

STATES OF CEE

Constantina Kottaridi?

Fragkiskos Filippaios

Marina Papanastassiou?

Athens University of Economics and Business, Department of International and European Economic Studies, Athens, Greece

?The author is thankful to the “Foundation Propontis” for its financial support. ?Assistant Professor and Visiting Scholar at the University of Reading. Author to whom correspondence should be addressed. Athens University of Economics and Business, Department of International and European Economic Studies, 76 Patission Str., 104 34, Athens, Greece. Tel: 003-010-8203711, e-mail: [email protected]


2

Abstract In a globalized environment, where intense competitive pressures constitute the

main features, Central and Eastern Europe is emerging as a promising investment

host region. Following the main lines of Vernon’s Product Life Cycle (PLC),

Hirsch’s International Trade and Investment Theory of the Firm (ITITF) and

Dunning’s Investment Development Path (IDP), a theoretical argumentation is

synthesized and basic hypotheses are posed and empirically tested. Results

demonstrate that these economies are going through the second stage of IDP and

confirm a complementary relationship between FDI and Imports as well as the

importance of specific location advantages as key determinants of inward FDI in

the region.

Keywords: CEE, product cycle, investment development path, inward FDI, imports. JEL Classification: F210 F230 F290 M110 O520


3

1. Introduction

In a world characterized by an ongoing degree of globalization where accelerated

technological progress, new production, organizational and management systems and a

constantly growing role of competition constitute the main features, it is imperative for

countries and enterprises to be internationally competitive in order to survive and grow.

The key challenge facing countries, in particular developing ones, is how to meet intense

global competitive pressures while sustaining growth. At the same time, enterprises need to

develop elaborated corporate systems in response to economic and technological forces.

It is in this global setting that Central and Eastern European (CEE) countries have emerged as

an important locus for foreign investors. The issue is of particular significance as Foreign

Direct Investment (FDI) is considered to be a vehicle through which new technologies and

knowledge are transferred. These, in their turn, affect the production functions of firms and

help boost the economy, contributing further to their transformation from centrally planned to

open-market economies and consequently, to their convergence with the European Union

(EU)-15.

It is interesting to mention that although global FDI inflows declined by more than 40%

following the global economic slowdown, flows into this region grew by 2% in 2001, while

it’s share of world inflows rose from 2% in 2000 to 3.7% in 2001 (World Investment Report,

2002). The above mentioned stylized facts indicate that CEE is considered a stable and

promising location for FDI, strengthened by the fact that its overall economic growth has

been affected less by the global economic slowdown than any other region.

Whilst the above hold for the region as a whole, the distribution of FDI is uneven among the

countries due to their different transition progress. The vast majority of FDI has been received

by the Czech Republic, Hungary and Poland, which are the first to begin liberalization and the

largest among the region, although during the time span, Hungary, the Czech Republic and

Estonia have had high inflows relative to GDP for some years, whereas Poland and Latvia


4

have been experiencing growing inward investment only recently (Holland and Pain, 1998).

On the contrary, Bulgaria and Romania receive much lower levels of FDI due to their

relatively poor progress in meeting the economic conditions for their accession to the EU

(Bevan et al., 2001).

Insert Table 1 here

Major investors in the region are Western European countries, especially Austria and

Germany, and the US. Austria has a special relationship with the region based on personal ties

and links and this is clearly illustrated by the fact that of total Austrian new direct investment

abroad, nearly 40% were allocated in CEE countries in 2000, a figure which rose to 80% in

2001 (Hunya, 2002). Germany too, has traditionally had close ties of culture, tradition and

language, as well as tight economic integration with the countries of CEE (Holland and Pain,

1998). US investors are registered to hold a significant role in these states especially in Poland

where more than 30% of capital is of American origin (Meyer, 1998).

The present paper focuses on the new EU member-states of CEE1. To date, investigation of

the CEE region has still been scarce due to data limitations. The studies that have already

been carried out are of limited time horizon for which no determinate conclusions could be

drawn. We instead use an extended time span dataset including even the more recent years up

to 2000, which enables us to explore the FDI motivations during the whole period of nineties.

Building on Vernon’s Product Life Cycle (PLC) Theory (1966, 1979), Hirsch’s International

Trade and Investment Theory of the Firm (ITITF) (1976) and Dunning’s Investment

Development Path (IDP) (1993), the paper presents an integrated theoretical approach

placing emphasis on two issues: On the location determinants of inward FDI in these

countries and on the issue of the interrelationship between inward FDI and imports, i.e.

complementarity vs. substitutability. Following the argumentation, the econometric

specification uses the three-stage least squares technique developed by Zellner and Theil


5

(1962). This allows us to draw conclusions on both FDI and exports’ driving forces as well as

on the existing link between the two.

2. Theoretical background and argumentation development

2.1 PLC, ITITF and IDP

In this section we depict our theoretical argumentation in regards to the location determinants

of FDI and to the inward FDI - Imports relationship in the EU accessed countries of CEE plus

Bulgaria and Romania.

We build mainly on the grounds of three previously established theoretical works, the PLC

Theory, Hirchs’s International Trade and Investment Theory and Dunning’s IDP. These

approaches provide plausible explanations for the relation of inward FDI and imports whilst

they also describe their time pattern.

Let us consider the PLC first. The relation between outward FDI and exports is a function of

two particular elements: the nature of the product and the development status of the country.

Vernon, (1966), distinguishes among three stages of product development 2: New product,

Maturing product and, finally, Standardized product. During the first stage the product is

produced in a single advanced country where “communication between the market and the

executives directly concerned with the new product is swift and easy, and in which a wide

variety of potential types of inputs that might be needed by the production unit are easily

come by” (Vernon, 1966). This first stage is followed by the “maturing product” stage where

a certain degree of standardization both in the production process and in the characteristics of

the good takes place, whilst the need for flexibility declines. During this second stage some

foreign production of the product begins. Finally, the product enters the last stage of its “life”

becoming fully standardized. It is at this particular stage that production can take place even

in less developed economies, which can now offer comparative advantages in terms basically

of cheap labor inputs.


6

Hirsch (1976), acknowledging the relevance of PLC, introduces three variables in order to

capture firm-specific revenue-producing factors and information, communication and

transaction costs, which, together with input costs, provide useful insights for international

trade and investment behavior of firms. In particular, he poses his argumentation on the

existence of firm-specific know-how and other intangible assets (K), the export-marketing

differential based on international transactions costs (M) and costs of controlling a foreign

subsidiary versus a domestic plant (C). His conditions thus for the invest versus export option

are presented as follows:

Firm in country A will export in country B if:

1.2.

b

b

P M P KP M P C

?

?

? ?? ?

??

Firm in country A will invest in country B if:

3.4.

b b

b

P C P KP C P M?

? ?? ?

??

where Pa and Pb denote production costs in country A and B respectively3.

He then extends his basic model to include intermediate goods production and/or more goods

to account for the export-enhancing or export-replacing nature of direct investment in his

post-investment strategy. In a recent paper Almor et al. (2003) built further on the outward

FDI and exports relation -combined with Dunning’s OLI (Ownership, Location,

Internalization) paradigm- and argue that when location and ownership advantages occur in

different countries then ownership advantage is realized through FDI. Otherwise, ownership

advantage is realized through trade (p. 10).

However, this is less than the full picture. There is still something missing and this is the

dynamic aspect. Development process of a country is not independent of inward FDI.

Transfers of new technology and new ways of production enhance the growth and the

development process of the host economy. This process is described originally in Dunning


7

(1993) and later by Dunning and Narula’s IDP (1997) and Narula and Dunning (2000). IDP

is an application of the OLI paradigm to explain the changing level and pattern of the MNCs’

activity and its interaction with a country’s investment path. This theory fills in the puzzle by

posing the missing dynamic element into the theory of international production. IDP in its

updated version describes five stages of development 4 through which countries are

distinguished by their propensity to be outward and/or inward investors. According to this, the

CEE economies fall into the second stage of development, where domestic markets may have

grown either in size or in purchasing power5. This, in turn, induces undertaking of some local

production by foreign investors as a viable and profitable alternative. The L advantages of the

potential host6 are decisive especially for export-oriented industries, which basically exploit

natural resources and primary commodities, creating forward vertical structures in their

production into labor- intensive low technology and light manufactures (Dunning and Narula,

1997; Venables, 1999).

Figure 1 presents the evolution of the IPD for the eight new EU members plus Bulgaria and

Romania during the period 1992-2000. It is evident that these countries belong to the second

stage of the IDP moving gradually to the third stage 7.

Insert Figure 1 here

In this sense, Location is emerging as indeed the “neglected” factor in our understanding of

FDI (Dunning, 2003). In particular, Dunning concludes “With the gradual geographic

dispersion of created assets, and as firms become more multinational by deepening or

widening their cross-border value chains, then, ..., the structure and content of the location

portfolio of firms becomes more critical to their global competitive positions” (p. 63)

2.2 Theoretical development


8

The trend during the last two decades has been the gradual move away from price competition

to differentiated goods competition, hence the emergence of multi-product firms producing a

number of varieties as dominant players in the international scene 8.

For the development of our framework, besides the three ‘cornerstones’ in the literature of

MNCs developed above, we borrow useful insights from Malley and Moutos, (1994) and

Baldwin and Ottaviano, (2001). We assume that a firm produces n final, differentiated goods

(n varieties or alternatively products), which are traded internationally. Each product is

produced in a single plant with declining average costs and its production function takes a

Leontief form:

min[ , ]i i i iQ L K S? ?? (1)

where Qi is the quantity of goods (varieties) produced, L and K stand for labor and capital

respectively, and S is a specific factor necessary to the production of goods.

Hence, we hypothesize that for the production of each variety a minimum combination of

labor and capital are required, plus the specialized factor S, where

( , )iS k CDS tcs?

CDS is the Country Development Stage and tcs is the trade cost9,10 of this specific factor.

Hence, the specific factor S is directly related to the stage of IDP of a country. We restrict

our argumentation in assuming that production of n depends on the location and availability

of S. In support of this assumption, Dunning also acknowledges that, in the 1990s, the market

-seeking type of FDI is influenced by a set of different location variables compared to those in

the 1970s. Amongst them he distinguishes the “availability and price of skilled and

professional labour” (Dunning, 2003, table 3.1, p.54).

Incorporating all the above assumptions to the standard PLC model, we result in the following

three production phases:


9

1. New Product

0i

i

QS

??

?,

2

20i

i

QS

??

? where i= new product

In this phase production is extremely sensitive to the existence of factor S, so that production

occurs only domestically. Thus, there is no FDI and only exports of the n products take place

(Almor, et. al, 2003). So,

1

( )n

i Ei

Q?

? › 0

where S stands for summation and the subscript E denotes exports.

In countries that fall into the first stage of IDP this factor is non-existent, whilst in advanced

countries this factor is abundant. It follows then that production of the n varieties under

consideration can take place only in developed economies11.

2. Maturing Product

0i

i

QS

??

?,

2

20i

i

QS

??

? where i = mature product

As a specific degree of standardization takes place, the need for flexibility declines while

concern about production costs arises. In addition, S may be found at least to some extent

even in other advanced economies or advancing economies that are in a higher stage of

development, like our sample of countries (World Bank, 1999). The relation between FDI

and Exports may be a substitutable or a complementary one. Let’s elaborate further on the

FDI and Exports relation:

According to Baldwin and Ottaviano (2001), if varieties are sufficiently good substitutes,

shifting production of one variety abroad, say variety 1, puts in effect the ‘cannibalisation’

effect, i.e. leads to price and quantity changes that harm export sales of the un-shifted

varieties since these face more competition in the foreign market. So, here, from one hand,

exports of the shifted variety are replaced by local production (this holds in the case that the


10

shifted variety was previously exported) and, on the other hand, export sales of the other

varieties i = 2, …, n are also decreased due to competition.

1( ) 0EQ ? , 2

( )n

i Ei

Q?

??

Let’s consider now the case that varieties are not perceived as being close substitutes. Then

the above-discussed mechanism of ‘cannibalization’ would not be generated, since respective

demand of un-shifted varieties still exists. In fact, if some of the goods are complements to a

degree, shifting production of one or some varieties abroad may lead to increased exports of

the other varieties. This may happen because due to local catering of the foreign market,

consumers’ cost is lower, thus their quantity demanded would be higher. By demanding a

greater quantity of variety i for example, automatically, quantity demanded for its

complements would rise too, so, export sales of complementary varieties would increase.

Assuming again that production of variety 1 is shifted:

1( ) 0EQ ? , 2

( )n

i Ei

Q?

??

So, in this case a complementary relationship is expected.

However, incorporating Hirsch’s (1976) post investment strategy argument, taking into

consideration previously not exported goods (varieties), the picture may be altered. Consider

that in addition to the exported varieties i = 1, …, n, the firm also produces varieties n + 1, n

+ 2, …, N, which due to their high export-marketing cost M are not exportable. Hirsch (1976)

argues that by establishing an affiliate abroad for the production of some variety, “the firm

changes the export perspectives of goods n + 1, n + 2, …, N which were previously not

exportable…Thus, even if international investment results in the elimination of certain

exports, it turns into exportable other goods…” (p. 265). Thus, the two cases above are

expressed below as:

1( ) 0EQ ? , 2

( )n

i Ei

Q?

?? /2

( )n

i Ei

Q?

?? , 1

( ) 0N

i Ei n

Q? ?? ?


11

Since we are talking about differentiated production, a crucial hypothesis for firms to survive

in the long run is their innovative capabilities in continuously producing new goods. So, it is

sensible to believe that by the time variety 1 is matured and, consequently, relocated, the firm

has already introduced a new product N+1. Malley and Moutos, (1994), borrow Krugman’s

(1979) insight and define the rate at which this relocation takes place as the ratio at which

new goods are produced. In this case then:

1( ) 0EQ ? , 2

( )n

i Ei

Q?

?? /2

( )n

i Ei

Q?

?? , 1

( ) 0N

i Ei n

Q? ?? ? , 1( ) 0N EQ ? ?

The analysis above indicates that there are strong reasons to expect that a complementary

relationship between foreign production and exports holds.

3. Standardized Product

0i

i

QS

? ?? ? where i=standardized product

In this phase investors are primarily concerned with low-cost captive markets, whilst

standardized product and process characteristics are put forth. Nonetheless, products still

require a particular amount of the specific factor S in order to be produced. Hence, by the

definition of this phase, there is no decision of the firm to establish an affiliate in an advanced

economy. Its decision relates solely to selecting in which of the low-cost countries will move

the production of these products. Exports of standardized products are replaced by foreign

production.

Consider again that good (variety) 1 is standardized, thus shifted to the less developed

country. Then, the following holds:

1( ) 0EQ ? , 2

( )n

i Ei

Q?

??


12

3. Related literature review

This section reviews the related literature on FDI in CEE countries carried out to date. It

should be mentioned here though that none of these studies applies a similar to the present

paper theoretical approach in the explanation of FDI determinants in the region. Nevertheless,

we feel it is appropriate to describe them as we will later on incorporate some of the variables

discussed in our empirical evaluation of the main hypotheses stated in the following section.

Generally speaking, there is a similarity between the relevant literature on the transition

economies and that on less developed countries (LDCs). Both lines of research distinguish the

market-seeking and efficiency-seeking motives as the key features of investing abroad

(Dunning, 2003). Lecraw’s (1991) analysis of GNP-normalized inflows in LDCs during

1974-86 concluded that investments oriented towards domestic markets are largely affected

by the growth of domestic demand, whereas labor costs adjusted for quality seem to affect

only the export-oriented investments. Caves (1996) reports that MNCs in these countries fall

into two categories; either producing primarily for exports or catering the local market and

states that they are mostly active in sectors that undertake labor-intensive stages of

processing.

Moving to the countries under consideration, it is worth mentioning that these belong to a

different economic process characterized by their transition from a centrally planned political

system to a market-oriented one in the light of their accession to the EU.

Specifically, the analysis of the motives that drive foreign capital into the transition

economies builds on the work of Lucas (1993) and Jun and Singh (1996), whose focus is on

the business environment, trade integration, labor costs and the form of privatization process.

Most authors seem to agree that, as in other developing economies, political and economic

factors have a key role to play in attracting FDI across the region.


13

Large markets are considered a significant determinant of FDI from all relevant FDI theories

and have a prominent role to play in Dunning’s (1993) market seeking behavior of firms 12.

In what regards the CEE markets, they are potentially attractive to many Western European

investors, who search for new opportunities for expansion (Lankes and Venables, 1996).

Consumers in CEE countries had no access to many consumer goods that were available to

people at similar per capita incomes. For consumer durables and manufactures of fast-moving

consumer goods, trade liberalization opened growth opportunities for firms whose established

markets in the West were saturated (Meyer, 1998). Boeri/Brücker et al., (2000), claim that

quite often an important investment motive is to supply the markets of the FDI host country

and to exploit the first-movers advantages in markets with no or limited competition. Market

factors, and in particular, “capturing local market” and “enlarging market share” were

reported as the primary motive for FDI in Pye’s investigation of major western investors in

Central Europe and in Rojec and Jaklic (2002) in their survey for Slovenia respectively.

Bevan and Estrin (2000), provide sound empirical support for the market size hypothesis too

(also, Meyer, 1998 and Altomonte and Guagliano, 2001).

Cost factors play a significant role in the classical efficiency- seeking behavior of

investors13Dunning (1993). Wage differentials are obviously more relevant for the CEE

countries since labor is regarded as having relatively high levels of skills and training

compared to other developing economies, whilst, although unit labor costs have risen

considerably, they are still far below West European, especially German levels. Labor costs

have been found to exert a significant effect on foreign investments in the region, either when

examining solely the wages (Holland and Pain, 1998) or even when taking labor productivity

into consideration (Bevan and Estrin, 2000)14. Nonetheless, it is argued that FDI into the CEE

countries are not motivated exclusively by low labor costs but to a great extent, in relation to

the market access motive (Meyer, 1998, Boeri/Brücker et al., 2000). However, taking


14

advantage of the wage differentials may be an important factor for at least some tradable

sectors, given the high labor intensity of FDI into the region (Weise et al., 2001).

The decision to engage in multinational activity into developing economies is directly linked

to the risk of the environment under consideration and much more to its macroeconomic

stability. Uncertainty with regards to macroeconomic conditions, for example, the interest

rate, the inflation rate, the exchange rate, the government balance, the external debt stock or

the end of year reserves, inhibits the undertaking of FDI in these economies and drives

investors to locations that are credit-worth. Macroeconomic stability can be measured in a

number of ways and each investor has a different perception of it, but in general, we would

argue that the above-mentioned measures may very well indicate how stable an economy is.

Performing a principal components analysis, both studies of Holland and Pain (1998) and

Bevan and Estrin (2000) for the periods 1992-1996 and 1994-1998 respectively, identify

several factors, i.e. interest rates, industrial development, government balance, inflation and

corruption among others, as determining the perceived country risk of their samples and

exerting a significant effect on the region’s inward FDI.

Geographical proximity to other markets is argued to affect inward FDI positively on the

grounds of existing business linkages and knowledge of the markets (Brainard, 1997). In

addition, within the international production systems of MNCs which seek to tap into global

value chains and take advantage of the new markets and lower labor costs, producers of

advanced economies may undertake production in developing countries positioned close to

them, with the aim of exporting back goods to home and other markets.

Meyer (1998) and Ebbers and Todeva (1999), mention the increased attractiveness of

Hungary, Czeck Republic and Slovak Republic to Austrian and German investors, as well as

the attractiveness of the Baltic states to the Nordic countries (also Boeri/Brücker et al., 2000).

These, i.e. Latvia, Lithuania and Estonia, are of particular interest because, though

geographically distant from most potential investors, they are psychologically much closer as


15

their traditions, languages and institutions are linked to the Baltic basin, primarily

Scandinavia (Bevan and Estrin, 2000).

In accordance with the above, exports have been tested in the literature in order to explore

their role in creating linkages with CEE markets. It is also argued that insofar as FDI is

motivated by market access -regarding the tradable sectors- it is complementing rather than

substituting trade, i.e. it raises the value added of parent companies in home countries relative

to a case without foreign investments (German Institute for Economic Research and EPRC,

2001). A direct test of the argument is carried out in Brenton, Di Mauro and Lücke, (1998),

the gravity model of whom provides strong empirical evidence of a complementary

relationship. Although there have been numerous studies trying to assess the relationship of

FDI-Exports (Lipsey and Weiss, 1981; Narula and Wakelin, 1999; Lipsey et al., 2000) for

advanced economies, no such work exists for the countries under consideration. This paper,

thus, fills the gap in the existing literature by attempting to clarify the above relationship

within the integrated context developed earlier.

Within the framework of intense competitive forces arising from globalization, even labor-

intensive activities do not any more search for unskilled workforce, but rather, processes

require some technical and creative abilities, an argument consistent with new trade and new

economic geography theories which consider skilled labor as an important agglomeration

factor. As education levels rise from very low levels, this would initially be perceived as an

improving location-advantage, hence would attract MNCs offering them a better environment

for the use of their existing ownership-advantages (Papanastassiou and Pearce, 1991). From

then onwards, the existence of high educational levels signals the ability of domestic labor

force to support competitive and developed production procedures. In addition, it is argued

that foreign firms use more capital- intensive techniques, thus, they require more skilled labor

and it is rather the quality than the cost aspect of labor that they are interested in (Rojec and

Jaklik, 2002).


16

Finally, it has been proved that announcements regarding the EU accession progress of the

countries under consideration have significant implications for the region’s inward FDI

(Bevan and Estrin, 2000).

4. Hypotheses, sample description and econometric specification

In the second section we argued that the countries under investigation are classified as stage 2

countries according to the IDP. This was revealed by the calculation of the IDP coefficient as

it is presented in Figure 1. Thus, two are the major hypotheses (H) to be tested in line with

our theoretical argumentation: The first hypothesis concerns the relationship between FDI

and trade and is formulated as follows:

H1: There is a complementary relationship between inward FDI and imports for countries

that belong to the second stage of the IDP.

The second hypothesis concerns the determinants of inward FDI in second stage IDP

countries. Building again on the theoretical specification of the mature product phase as this

is analysed earlier in this paper and on Dunning and Narula (1997) and Narula and Dunning

(2000), the second hypothesis is formed as follows:

H2: The existence of advanced location characteristics, i.e. the availability of S factors, in

second IDP stage countries, will exert a positive influence on inward FDI.

Our sample includes the eight new members of the enlarged EU, i.e. Czech Republic, Estonia,

Hungary, Latvia, Lithuania, Poland, Slovak Republic and Slovenia and the two candidate

countries, i.e. Bulgaria and Romania. The time period covered extends from the early


17

transition stages in 1992 till 2000 (for the descriptive statistics of the sample see Table 1 in

Appendix 1). Our sample has a structure of a balanced panel dataset.

An approximation of the production function (1) and in order to test for our hypotheses, the

following system of equations is examined:

1 2 3 5 6 7 8

9 10 1

1 2 3 4 5 2

it it it it it

it

it it it it it it

FDI IMP GDP IRS WG SECit aTERTit

a CPit a ENLRGit

IMP FDI GDP GDPC OPEN

? ? ? ? ? ??

? ? ? ? ? ?

? ? ? ? ? ? ? ?? ?

? ? ? ? ? ?

(1)

where i stands for the country under consideration and t for the respective year.

The FDI equation is a function of imports (IMP). FDI is measured both as flows and stocks.

This is because stocks, contrary to flows, reflect the “embededdeness“ of foreign investors in

local markets. Consistent with the aforementioned existing literature on FDI determinants, we

included in the FDI equation Gross Domestic Product (GDP) as a measurement of the

“market- seeking behavior” as this constitutes a strong characteristic in the second stage of

IDP and consequently we expect a positive relation. Labor costs according to our production

function (1) are measured by wages (WG)) reflecting a more traditional “efficiency-seeking

behavior”. Thus, a negative sign agrees with the location characteristics expected for second

stage IDP. The interest rate spread (IRS)15 is included as a measure for the risk premium of

the economy on the grounds that a higher difference implies a non-credible and non-stable

market. The specific factor S of our production function (1) is captured by the sophistication

of the labor force as indicated by enrollments in secondary and tertiary ratios of the relevant

population age (SEC and TERT respectively16). In this case a positive sign will be in support

of the emerging new patterns of specialized location determinants. To adjust for country and

time effects we included the following dummies: for cultural proximity of the Baltic States to

the Nordic countries we implore a dummy variable, CP as FDI also is assumed to differ in the


18

Baltic region firstly due to their being far away from the rest of Europe and secondly because

they receive flows from particular origins due to cultural factors. FDI flows are also

considered to be influenced by the reaffirmation of the Madrid European Council about EU’s

enlargement in 1995, thus, we incorporate a time dummy variable from 1995 onwards,

namely, ENLARG.

The imports equation is basically standard, according to which imports are affected positively

by the GDP of the importing country (GDP), GDP per capita (GDPC), the openness of the

underlying economy (OPEN) measured by the share of total trade to the respective country’s

GDP and the level of foreign investments taking place in line with our theoretical analys is

(FDI). Particular attention should be paid on GDPC because this indicates the respective

development level of a country and, hence, its needs for imports. Data were collected from the

European Bank for Reconstruction and Development (EBRD), Internationa l Financial

Statistics (IFS), International Labor Office (ILO), United Nations for Cooperation on Trade

and Development (UNCTAD) and World Bank (WDI) electronic databases and various

issues.

In matrix notation, our system of equations is written:

1 2 1 1 1 1

2 1 2 2 2 2

0 00 0

Y Y XY Y X

? ? ?? ? ?

? ? ? ?? ? ? ?? ? ? ?? ? ?? ? ? ?? ? ? ?? ? ? ?

? ? ? ?? ? ? ?? ? ? ?

where Y1 represents inward FDI, Y2 stands for imports and are the dependent variables. X1

and X2 are a set of explanatory variables.

Ìn line with the theoretical modelling, i.e., on the existence of either substitutable or

complementary relationship between inward FDI and imports, we proceed with Zellner and

Theil (1962) 3SLS (Three-stage Least Squares) estimator to get consistent and efficient

estimators of our system in order to account for endogeneity. The 3SLS satisfies the

requirements for an IV (Instrumental Variable) estimator and therefore is consistent. The IV


19

in this case makes use of the equation correlations of the disturbances. Furthermore among

all IV estimators that use only the sample information embodied in the system, 3SLS is

asymptotically efficient17.

A relevant question that arises when using panel data regards the poolability or not of the

sample. An established method to investigate the issue is to test for the pooled vs. fixed

effects techniques in the single equation of FDI. (See Appendix 2 for a detailed technical

analysis. Relevant tests and criteria are also calculated in Appendix 2. Both the restricted and

the unrestricted models are illustrated in Table 1 in Appendix 2)).

In order to further support the choice of the pooled model, we proceeded to the within-groups

estimator in the system of equations (1), which can be calculated as follows:

. . .( )it i it i it iy y x x? ? ??? ? ? ? ?

Where the group means are:

. . .i i iy a x? ??? ? ?

This within-groups estimation is directly comparable to the fixed -effects one, (Least Square

Dummy Variable (LSDV) estimation) but does not calculate the individual effects, allowing,

thus, for greater flexibility in terms of degrees of freedom. (Table 2 in Appendix 2 provides

both the pooled and the within-groups estimation methods and demonstrates the robustness of

the results).

4. Results

Results are presented in tables 2 and 318. At a first stage, we estimate the model without the

dummy variables, i.e. without controlling for the cultural proximity of the Baltic states with

some countries and the impact of the enlargement reaffirmation in 1995 (Eq.2.1 and 3.1). At

later stages, we include the Baltic dummy, the enlargement dummy and both dummies at the

same time respectively.

Insert Table 2 and 3 here12


20

Overall the results confirm our hypotheses about the complementary relationship between

inward FDI and imports (i.e. Eq 2.1 and 3.1) and thus justifying our simultaneous equations

technique. More analytically and starting with table 2 we observe:

GDP and WG turn out to exert a great impact on FDI flows confirming both the market-

seeking as well as the traditional “efficiency- seeking” motive developed for these type of

countries. The secondary enrollment ratio (SEC), which stands for medium workforce

capabilities and skills, provides a positive and statistically significant sign in two out of four

equations in the FDI flows estimation whilst the tertiary enrollment ratio, representing high

education and expertise appears to be negative. At the same time, in the imports equation,

GDP per capita gives a very strong positive impact. GDP per capita may be perceived as

having demand-side and supply-side influences and the way it affects imports may alter as it

changes. On one hand, taken as a measure of consumer taste, it indicates the appropriateness

of foreign production for host country markets, whilst on the supply-side GDP per capita

‘may well be related to the technological and managerial capability of the host country’

(Papanastassiou and Pearce, 1991). Rising GDP per capita is expected to affect imports

positively on the grounds that consumers desire more advanced and qualitative products

produced elsewhere in the world, which local industrial structures cannot support (Linder

effect, Linder 1961).

Combining these results, it is evident that FDI is directed to Central and Eastern Europe

mainly to take advantage of the new markets producing goods and to exploit the cheap cost of

labor. Our results are consistent with the above even when we add the cultural proximity

dummy to the model, which, nevertheless, doesn’t indicate any particular effect (Eq. 2.2).

A slight differentiation emerges when the announcement of the EU enlargement is included

(Eq. 2.3). The variable seems to have influenced to a great extent inward foreign investments

in the region, which is actually expected. Reaffirming commitment to EU enlargement is


21

really an indication that all countries have been doing well in their development hence, it is a

reassurance for a stable and not risky any more environment. In addition, the perspective of

being included in the EU acts as an ‘umbrella’ for the region offering it protection and

prospects for the future. Our SEC variable turns out to be significant now confirming the fact

that medium workforce capabilities are mostly required for the production of the particular

goods. In this model, the OPEN variable in the imports equation, is also affecting imports

significantly, a normal outcome for our imports equation.

Results in table 3 provide a very strong support for our second hypothesis, i.e. that one of the

major location determinants of inward FDI is the availability of the S factor. The consistently

strong positive sign for SEC reveals that existing investors have a deeper insight of the local

economy. This is reflected by the gradual departure from cheap labor considerations (as

expressed in wages) towards the search of more specialized inputs.

Finally, in order to answer the question on causality, we instrumented FDI and Imports by

applying one year lag and we found that imports (as probably expected) determine FDI and

not the other way around 19.

5. Concluding Remarks

Central and Eastern Europe has recently been a hot topic of discussion within both academic

and political circles. The developments in the region have, not surprisingly, raised interest

from the part of foreign investors who now detect viable and profitable opportunities.

This paper develops a novel approach to the analysis of inward foreign investments in the

region, combining elements from Vernon’s Product Life Cycle Theory, Hirsch’s International

Trade and Investment Theory of the Firm and Dunning’s Investment Development Path.

Following the main lines of these theories, we build a theoretical argumentation based on a

Leontief production function in which we specify the region’s development stage and the type

of foreign activity attracted and, consequently, the relationship between Inward FDI and


22

Imports. Our results suggest that these economies are going through the second stage of the

IDP. This stage of development is associated with the production/export of mature products

to these economies. In the case of FDI strong locational characteristics such as market size

and quality of labor play a positive role. In regards to the inward FDI vs. Imports relationship,

results strongly indicate complementarity both when examining FDI inflows and FDI inward

stock. This raises the issue of core-periphery within an enlarged EU. However, it is too soon

to address it properly due to the recent adhesion date of these countries.

Further research, may focus on industrial level analysis which for the time being is beyond

reach due to lack of reliable and consistent data.


23

Table 1. Inward FDI for the analysed CEE countries (in millions of USD) Country 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Bulgaria 42 40 105 90 109 505 537 819 1002 689 Czech Republic 1003 654 869 2562 1428 1300 3718 6324 4986 4916 Estonia 82 162 215 202 150 267 581 305 387 538 Hungary 1471 2339 1146 4453 2275 2173 2036 1944 1643 2414 Latvia 29 44 213 178 382 521 357 348 408 201 Lithuania 10 30 31 73 152 355 926 486 379 446 Poland 678 1715 1875 3659 4498 4908 6365 7270 9342 8830 Romania 77 94 341 419 263 1215 2031 1041 1025 1137 Slovak Republic 100 168 245 195 251 220 684 390 2075 1475 Slovenia 111 113 128 177 194 375 248 181 176 442 Source: UNCTAD, on-line statistics


24

Table 2. Econometric Results of various models tested for inward FDI flows

(2.1) (2.2) (2.3) (2.4) FDI IMP 0.0818*** 0.0864*** 0.0713*** 0.0807***

4.43 4.56 3.57 4.08 GDP 0.0108* 0.0104* 0.0128** 0.0118**

1.86 1.82 2.14 2.04 IRS -1.0667 -1.2417 -0.2908 -0.4696

-0.23 -0.27 -0.06 -0.09 WG -0.1747*** -0.1484* -0.1577** -0.1139

-2.67 -2.04 -2.46 -1.53 SEC 35.8909 27.4437 38.6341* 21.5837

1.52 1.1 1.7 0.85 TERT -5.2980 -4.2012 -26.2327 -23.3407

-0.26 -0.21 -1.06 -0.98 CP 377.7651 605.1908

0.73 1.11 ENLARG 957.1861** 942.2606**

2.18 2.25 C -2691.7890 -2376.2620 -2931.3610* -1.56 -1.4 -1.81

Obs 83 83 83 83 R-sqr 0.77 0.77 0.79 0.79

Chi-sqr 285.72*** 295.75*** 309.3*** 327.6***

IMP FDI 5.7015*** 5.4957** 4.6823*** 4.6789***

2.87 3.01 3.81 4.18 GDP 0.0923 0.1001 0.1294 0.1301***

1.28 1.5 2.82 3.09 GDPC 2.2027*** 2.2334** 2.3457*** 2.3406***

4.64 4.92 6.11 6.29 OPEN 169.0603 192.6480 209.4219* 233.3387**

1.53 1.7 1.91 2.11 C -13368.7000* -14812.7700 -16505.8800** -17711.6400*** -1.87 -2.04 -2.45 -2.62

Obs 83 83 83 83 R-sqr 0.87 0.88 0.89 0.89

Chi-sqr 569.1*** 599.24*** 722.14*** 746.19*** ?Coefficients and z-stat respectively *** Significant at 1% ** Significant at 5% * Significant at 10%


25

Table 3. Econometric Results of various models tested for inward FDI stock

(3.1) (3.2) (3.3) (3.4) FDISTOCK

IMP 0.1845*** 0.1865*** 0.1655*** 0.1776*** 4.18 4.01 3.76 3.96

GDP -0.0025 -0.0021 0.0006 -0.0009 -0.18 -0.16 0.05 -0.07

IRS -1.0302 -1.0021 2.7053 2.5308 -0.08 -0.08 0.22 0.21

WG -0.2238 -0.1935 -0.1819 -0.1364 -1.52 -1.07 -1.29 -0.8

SEC 156.6566*** 148.3438*** 144.9072*** 123.7952** 2.87 2.25 2.84 2.05

TERT 1.9291 3.6537 -59.3240 -56.0816 0.04 0.07 -1.08 -1.03

CP 334.3444 689.5904 0.25 0.55

ENLARG 2534.1530*** 2490.8380*** 2.69 2.68

C -13450.4800*** -13151.3800*** -12349.8900*** -11315.2200*** -3.40 -3.08 -3.36 -2.90

Obs 83 83 83 83 R-sqr 0.72 0.72 0.75 0.75

Chi-sqr 204.44*** 203.90*** 234.30*** 237.44***

IMP FDISTOCK 1.6648*** 1.6578*** 1.6474*** 1.6795***

3.37 3.36 4.08 4.17 GDP 0.2048*** 0.2052*** 0.2061*** 0.2045***

6.94 6.97 8.21 8.15 GDPC 2.0711*** 2.0771*** 2.0834*** 2.0657***

4.99 5.02 5.44 5.39 OPEN 214.5328** 219.6172** 234.7961** 243.1621**

2.05 2.07 2.24 2.29 C -16190.2600*** -16488.9300*** -17310.7500*** -17641.2700*** -2.61 -2.63 -2.78 -2.80

Obs 83 83 83 83 R-sqr 0.91 0.92 0.91 0.91

Chi-sqr 872.25*** 875.52*** 882.86*** 877.19*** ?Coefficients and z-stat respectively *** Significant at 1% ** Significant at 5% * Significant at 10%


26

Figure 1. Investment Development Path


27

References

Almor, T., Hasai, N. and Hirsch S. 2003 Elements of an Integrated internationalisation model, Proceedings Academy of International Business, Annual Conference Monterey 2003. Altomonte C., and Guagliano C., 2001. Competing locations? Market potential and FDI in Central and Eastern Europe vs. the Mediterranean, Discussion Paper No. 108, LICOS, Centre for Transition Economies. Amiti, M., 1998a. New trade theories and industrial location in the EU: a survey of evidence, Oxford Review of Economic Policy, 14, 45-53. Baldwin, and Ottaviano, G., 2001. Multiproduct multinationals and reciprocal FDI dumping, Journal of International Economics 54, 429-448. Barrell, R. and Pain, N., 1999a. Trade restraints and Japanese direct investment flows, European Economic Review, 43, 29-45. Bevan, A.A. and Estrin, S., 2000. The determinants of foreign direct investment in transition economies, Centre for Economic Policy Research Discussion Paper DP2638, William Davidson Institute Working Paper No.342, University of Michigan, and Centre for New and Emerging Markets Discussion Paper No. 9, London Business School. Bevan, A.A., Estrin, S. and Grabbe, H., 2001. The impact of EU accession prospects on FDI inflows to central and eastern Europe, ESRC ‘One Europe or Several’ Programme Policy Paper 06/01, Sussex European Institute, University of Sussex. Boeri, T. and Brucker, H. et al.,2000. The impact of eastern enlargement on employment and labour markets in the EU member states, Report for European Commission’s employment and social affairs directorate, European Integration Consortium (DIW, CEPR, FIEF, IAS, IGIER), Berlin/Milan, Germany/Italy. Brainard, S.L., 1997. An empirical assessment of the proximity-concentration trade-off between multinationa l sales and trade, American Economic Review, 87, 520-544. Brenton, P., Di Mauro, F. and Lucke, M., 1998. Economic integration and FDI: an empirical analysis of foregn investment in the EU and in central and eastern Europe, Kiel Working Paper No. 890, Kiel Institute of World Economics, Kiel. Caves, R.E, 1996. Multinational enterprise and economic analysis, Cambridge University Press, Cambridge and New York. Clegg, J. , 1996. The United Kingdom : a par excellence two –way direct investor in Dunning, J and Narula, R. (eds) Foreign direct investment and governments : catalysts for economic restructuring, Routledge : London. Culem, C., 1988. The locational determinants of direct investments among industrialized countries, European Economic Review, 21, 885-904.


28

Dunning, J., 1993. Multinational enterprises and the global economy, New York: Addison Wesley. Dunning, J. 2003 Location and the multinational enterprise: a neglected factor?, in Brewer, T, Young, S. and Guisinger, S eds. The new economics analysis of Multinationals: An agenda for management, policy and research, Cheltenham: UK, pp.45-69. Dunning, J. and Narula, R., 1997. The investment development path revisited. Some emerging issues, in Dunning, J. and Narula, R. (eds.) Foreign direct investment and governments, catalysts for economic restructuring. Ebbers, H. and Todeva, E., 1999. Exploring the links between foreign direct investment, privatization policies, and transition reforms in central and eastern Europe, European international Business Academy, 25th Annual conference ‘International Business and the Global Services Economy’ Manchester. EBRD, 2000. Transition Report 2000, European Bank for Reconstruction and Development, London. EBRD, 2000. Transition Report Update May 2001, European Bank for Reconstruction and Development, London. Greene, W. H., 2000. Econometric Analysis, London: Prentice Hall International. Hardy, J., 1994. Eastern promise? Foreign investment in Poland, European Business Review, 94, 5, 28-37. Hirsch, S., 1976. An international trade and investment theory of the firm, Oxford Economic Papers, 28, 258-270. Holland, D. and Pain, N., 1998. The diffusion of innovations in Central and Eastern Europe: a study of the determinants and impact of foreign direct investment, NIESR Discussion Paper 137, National Institute of social and Economic research, London. Hunya, G., 2002. FDI in south-eastern Europe in the early 2000s, The Vienna Institute for International Economic Studies (WIIW), Vienna. IFS, 2000. International Financial Statistics Yearbook 2000,International Monetary Fund. IFS, 2001. International Financial Statistics August 2001, International Monetary Fund. ILO, 2000. Yearbook of Labour Statistics 2000, International Labour Organization, United Nations. Jun, K.W. and Singh, H., 1996. The determinants of foreign direct investment: new empirical evidence, Transnational Corporations, 5, 67-106.


29

Krugman, P., 1979. A model of innovation, technology transfer, and the world distribution of income, Journal of Political Economy87, 253-263. Krugman, P., 1980, Scale economies, product differentiation and the pattern of trade, American Economic Review, 70, 950-959. Lankes, H.P. and Venables, A.J., 1996. Foreign direct investment in economic transition: the changing pattern of investments, Economics of Transition, 4, 331-347. Lansbury, M., Pain, N. and Smidkova, K., 1996. Foreign direct investment in Central Europe since 1990; an econometric study, National Institute Economic Review, 156, 104-113. Lecraw, D.J., 1991. Transna tional corporations in host developing countries: a preliminary report, in Multinational Enterprises in Less Developed Countries, eds. Buckley, P.J. and Clegg, J. Linder, S.B (1961) An essay on trade and transformation, Wiley, New York. Lipsey, R., Ramstetter, E. and Blomstrom, M., 2000. Outward FDI and home country exports: Japan, the United States, and Sweden, SSE/EFI Working Paper Series in Economics and Finance No. 369. Lipsey, R. and Weiss, M., 1981. Foreign production and exports in manufacturing industries, The Review of Economics and Statistics 63 (4), 488-494. Lucas, R., 1993. On the determinants of direct foreign investment: evidence from east and southeast Asia, World Development, 21, 3, 391-406. Malley, J. and Moutos, T, 1994. A prototype macroeconomic model of foreign direct investment, Journal of Development Economics 43, 295-315. Markusen, J., 1984. Multinationals, multi-plant economies and the gains from trade, Journal of International Economics, 16, 205-226. Meyer, K., 1998. Direct investments in economies in transition, Edward Elgar. Meyer, K., 2001. Institutions, transaction costs, and entry mode choice in eastern Europe, Journal of International Business Studies, 32, 2, 357-367. Narula, R. and Dunning, J., 2000. Industrial deve lopment, globalisation and multinational enterprises: New realities for developing countries, Oxford Development Studies, 28, 2, 141-167. Narula, R. and Wakelin, K., 1998. Technological competitiveness, trade and foreign direct investment, Structural Change and Economic Dynamics, 9, 373-387. OECD, Main Economic Indicators, Organisation for Economic Cooperation and Development, various issues.


30

Papanastassiou, M. and Pearce, R. D., 1991/1992. Host-country determinants of UK FDI and exports: an analysis of developed and developing countries, Discussion Paper Series B, No.158, Department of Economics, University of Reading. Pye, R., 1998. Foreign direct investment in central europe : experiences of major western investors, European Management Journal, 16, 4, 378-389. Rojec, and Jaklik, , 2002. Integration of Slovenia into EU and global industrial networks: review of existing evidence, Working Paper No. 14, Centre for the Study of Economic and Social Change in Europe, School of Slavonic & East European Studies, University College London. Scetlicic, M. amd Jaklik, A., 2002. Outward FDI by selected Transition Economies; comparative evaluation, Competitive Paper, European International Business Academy (EIBA) 2002, Athens, Greece. Venables, A., J., 1999. Fragmentation and multinational production, European Economic Review, 43, 935-945. Vernon, R., 1966. International investment and international trade in the product cycle, Quarterly Journal of Economics, 80, 190-207. Vernon, R., 1979. The product cycle hypothesis in the new international environment, Oxford Bulletin of Economics and Statistics, 41, 255-267. Weise, C., Bachtler, J., Downes, R., McMaster, I. And Toepel, K., 2001. The impact of EU enlargement on cohesion, Final Report, European Commission Tender No. PO/00-1/RegioA4, DIW German Institute for Economic Research and European Policies Research Center, Berlin and Glasgow. World Bank (1999), Annual Report, Washington DC, World Bank Publications. World Development Indicators, World Investment Report, various issues, World Bank. World Investment Report, 2002. World Bank. Zellner, A. and Theil, H., 1962. Three stage least squares: simultaneous estimation of simultaneous equations, Econometrica, 30, 63-68.


31

APPENDIX 1. Descriptive statistics

Table 1. Statistics for the full sample.

Variable Obs Mean Std. Dev. Min Max

FDI 90 2498.922 3457.252 108.0104 15510.46 IMP 90 31744.61 28768.34 3700.152 121146 GDP 90 72009.17 78268.19 5807.566 326210.4 RP 83 15.05301 31.10814 -.3 269 WG 90 6552.117 3146.118 1999.856 14547.2 SEC 90 88.85624 10.08356 66 110.8861

TERT 90 28.62424 9.56181 11.8 59.8304 GDPC 90 6960.967 3126.858 3209.93 12981.74 OPEN 90 50.06686 8.450781 40.41967 78.37455

Collinearity Diagnostics

Variable VIF VIF^2 Tolerance Eigenvalue Condition Index

R-Square

GDP 1.55 1.25 0.6433 2.8298 1.0000 0.3567 GDPC 6.39 2.53 0.1566 1.3616 1.4416 0.8434 OPEN 1.70 1.30 0.5899 1.0763 1.6215 0.4101

RP 1.25 1.12 0.7986 0.7892 1.8936 0.2014 WG 5.72 2.39 0.1749 0.6157 2.1439 0.8251 SEC 2.27 1.51 0.4400 0.2460 3.3916 0.5600

TERT 2.38 1.54 0.4201 0.0814 5.8958 0.5799 Mean VIF

3.04 CI

5.8958


32

APPENDIX 2

The first question arising from the use of a panel dataset is whether to pool or not the

sample. The usual t-ratio for the individual effects ai tests for the hypothesis that

those are zero. In a regression context though this test is not useful in order to

determine whether to pool the data or not. To test for differences between groups we

can use an F-statistic. Under the null hypothesis, the efficient estimator is pooled

least squares. Coming back to the original equation for FDI flows the F-test between

the pooled equation and the one using individual effects, produces a value of

F(9,68)=1.19 which leads to the rejection of the fixed-effects model.

Following Greene (2000, p. 562) the F-test is estimated as: 2 2

2

( )/( 1)( 1, )

(1 )/( )u r

u

R R nF n nT n K

R nT n K? ?

? ? ? ?? ? ?

where n represents the number of cross-section elements, T stands for the time

periods, K is the number of coefficients and u denotes the unrestricted model while r

the restricted one (with only a single overall coefficient).

In addition to that, a Bayesian Information Criterion (BIC) was used to test the two

models, the pooled one and the one using the individual effects, producing a value of

17.046 which is a strong support in favor of the pooled model.


33

Table 1. Estimation results for the Unrestricted and Restricted Model. FDI is measured in flows.

Restricted Model Unrestricted Model FDI IMP 0.0763*** 0.0439

(4.45) (1.04) GDP 0.0137*** 0.0850***

(2.51) (2.74) IRS -5.3962 -3.9794

(-0.82) (-0.61) WG -0.0670 0.1445

(-0.92) (0.53) SEC 22.3492 -43.5513

(0.86) (-0.93) TERT 35.7629 53.0634

(1.44) (1.32) C -3373.7600* -15783.4300*** (-1.77) (-2.47)

Obs 83 83 R-sqr 0.7627 0.7952 F(6,78)=46.01*** F(15,69)=22.75*** BIC -84.87 -101.92

?Coefficients and t-stat respectively *** Significant at 1% ** Significant at 5% * Significant at 10%


34

Table 2: Results for the 3SLS pooled and the within-groups estimations for the system of equations.

FDI measured in flows

Pooled Estimation Within-Groups Estimation FDI (eq. 2.1) IMP 0.0818*** 0.0684***

(4.43) (4.08) GDP 0.0108* 0.0144***

(1.86) (3.11) IRS -1.0667 -10.4819

(-0.23) (-0.45) WG -0.1747*** -0.0992*

(-2.67) (-1.65) SEC 35.8909 6.5913

(1.52) (0.18) TERT -5.2980 18.5242

(-0.26) (0.57) C -2691.7890 -1022.9000 (-1.56) -0.36)

Obs 83 83 R-sqr 0.77 0.9684

Chi-sqr 285.72*** 395.66***

IMP FDI 5.7015*** 10.6024***

(2.87) (2.71) GDP 0.0923 -0.0674

(1.28) (-0.5) GDPC 2.2027*** 1.5885*

(4.64) (1.83) OPEN 169.0603 143.8610

(1.53) (0.58) C -13368.7000* -8153.8070 (-1.87) (-0.55)

Obs 83 83 R-sqr 0.87 0.9413

Chi-sqr 569.1*** 167.25*** ?Coefficients and z-stat respectively *** Significant at 1% ** Significant at 5% * Significant at 10%


35

Notes 1 The new EU member states are Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovak Republic, Slovenia. However, in our sample we have also included Bulgaria and Romania which were candidate countries although they didn’t manage to fulfil the accession criteria. These, though, do not alter our results. 2 In his subsequent paper in 1979, he argues that PLC may not be any more relevant when talking about innovations from US as other countries (like the western European ones) have closed the gap in terms of per capita income and innovative capabilities. However, he considers it relevant when talking about these very advanced economies and other developing ones. 3 Later on, Agmon and Hirsch, 1979, incorporate these basic lines to pose respective conditions with regards to less developed economies. 4 IDP, in its original form, distinguishes among four stages of economic development 5 Svetlicic and Jaklik, 2002 mention that outward FDI is a new phenomenon in these countries and only in 2000 it did grow faster than inward FDI. This supports the categorization as being in the second stage of IDP moving only recently and gradually to the third stage. 6 The L specific characteristics of a country (L standing for location) refer to location-bound capabilities and resources of a host, such as transportation, skilled and uskilled labor and communication facilities. L is one of the three elements of Dunning’s OLI Paradigm. 7 The relevant table with calculated IDP coefficients, following Clegg (1996) is available upon request. 8 Baldwin and Ottaviano (2001) mention the cases of Nestle and Procter and Gamble, which engage in the production of a vast range of consumer goods. 9 S resembles K in Hirsch’s formulation in that it may represent intangible assets, such as managerial know-how, technology, networks, special services, etc. as well as specialized intermediate inputs. 10 Although intangible assets in general are considered to display characteristics of a public good (Hirsch, 1976), we claim here that there are export costs for them too, since, for example, in order to use a specific process technique in a foreign country, executives would have to travel there and teach the personnel. We define trade cost as representing transport and communication costs plus coordination costs in the form of C in Hirsch, 1976. Thus, exports of S, SiE, would be: for new products 0iES ? , for maturing products 0iES ? and finally for standardised products 0iES ? but low. 11 In Malley and Moutos, 1994, these n goods (varieties) are distinguished between the ‘new’ and the ‘old’, which are both produced in advanced economies, because although local labour has the skills to produce them, host firms cannot produce them, for they lack the specialised factor required. 12 It is worth mentioning that large markets hold a particular role in new trade and new economic geography theories, as they reflect the potential of firms to capture scale economies (Krugman, 1980, Amiti, 1998a). 13 Cost factors are at the heart of the Hecksher-Ohlin traditional trade theory. Foreign investments are considered to be motivated by production cost differentials, which investors try to exploit in order to increase their profits by reducing their cost of production, however, explaining FDI determinants within this framework is beyond the scope of this paper. 14 See also Lansbury et al. (1996a, b), Meyer, (1998, 2001) and Hardy (1994). 15 IRS is calculated as the lending rate of host economy i minus its deposit rate. 16 These refer to gross enrolment ratios (as defined by the World Bank), i.e. the number of people (irrespective of their age) in the secondary and tertiary education over the relevant age groups respectively 17 See Schimdt (1976) for a detailed analysis of the 3SLS efficiency relative to 2SLS. 18 Collinearity diagnostics of the model are found in Appendix 2. Stationarity tests showed no evidence on the existence of a unit root. These results are available upon request. 19 Regressions with the lagged values are not included in the paper due to space limitations. They are nevertheless available upon request.

Filename: CEEJIBS1.DOC Directory: D:\Documents and Settings\sxs01ag\My

Documents\web Template: D:\Documents and Settings\sxs01ag\Application

Data\Microsoft\Templates\Normal.dot Title: 1 Subject: Author: PhD Deos Keywords: Comments: Creation Date: 27/02/2004 14:25 Change Number: 2 Last Saved On: 27/02/2004 14:25 Last Saved By: Anne Goodwyn Total Editing Time: 1 Minute Last Printed On: 27/02/2004 14:26 As of Last Complete Printing Number of Pages: 35 Number of Words: 7,976 (approx.) Number of Characters: 45,466 (approx.)