FDI and the Multinational Corporation:
New Theories and
Evidence
Training and Mobility of
Researchers Network
Research
1. Theoretical Modelling
In
the early period of the project, the Swedish team extended the
neoclassical growth model with FDI in a project that assessed the social
rate of return to FDI. They then conducted a panel data survey that
indicated support for the model’s predictions.
The
theory of the multinational corporation and international trade
Anthony
Venables, of the LSE team, has completed modelling work on the interaction
between trade and FDI, on multinationals and the vertical fragmentation of
production, and the impact of multinationals on the host economy. His 2000
paper with Jim Markusen (UPF and CEPR) won the Bhagwati prize for the best
article in the Journal of
International Economics, 1999-2000. Markusen and Venables have also
done work published in the paper ‘The International Organization of
Multi-Stage Production’, which was presented at the closing conference
of the network (see below).
Also
from the LSE team, Nico Matouschek worked on the boundaries of the firm,
investigating the firm’s decision between different organisational
structures of its foreign subsidiary (wholly-owned subsidiary, joint
venture, licensing). In another paper, Matouschek investigated the impact
of FDI projects on vertically-differentiated industries.
Studying
the choice of penetrating distant markets, the CSLA team followed two
directions of theoretical research. First, Basevi and Ottaviano modelled
the choice between exporting and FDI by Marshallian industrial districts
(MID). They argued that, by the very nature of the MID, its firms are
bound to resort to a combination of exports and FDIs that is inefficient
from the point of view of the district as a whole. Industrial districts
are depicted as centres of innovation in which local technological
externalities sustain the endogenous invention of new goods by
profit-seeking firms. After invention firms face a crucial choice between
reaching distant markets by export or plant delocation. The model shows
how firms, in an attempt to circumvent the obstacles to goods and plant
mobility, overlook the impact of their decisions on innovation activities
inside the district thus generating a suboptimal mix of export and
delocation. Second, Ottaviano and Turrini introduced incomplete
outsourcing contracts in an otherwise standard model of MNEs based on the
trade-off between proximity and concentration. They demonstrate that
incomplete outsourcing contracts can account for the observed emergence of
FDIs in large markets not only when trade costs are large but also when
trade costs are small. In addition, contractual incompleteness is showed
to alter someway dramatically the choice of supply mode made when
contracts are complete.
In
addition, analysing the relationship between relocation, product quality
and skilled employment, Barba Navaretti, Falzoni and Turrini developed a theoretical model on the choice of relocation for a
vertically (different product quality) and horizontally (different
varieties) differentiated industry, where firms are heterogeneous in their
ability to produce quality. The model predicts that vertical investment in
cheap labour countries will be undertaken by the firms that find greater
difficulties in adding quality to their products. This is because quality
requires skills that are scarce in low-wage, less-developed economies. At
equilibrium, it emerges that investments generate a more abundant flow of
intra-firm trade when they are located in cheap labour countries, and that
the firms with a small share of skilled workers in their parent company
are those that are more prone to invest in low-wage locations.
Holger
Görg (University of Ulster, SSE and University of Nottingham), F Walsh (UCD)
and Eric Strobl (UCD) worked on a project looking at why foreign-owned
firms pay more, looking in particular at the role of on-the-job training.
Foreign-owned firms have consistently been found to pay higher wages than
domestic firms to what appear to be equally productive workers in both
developed and developing countries alike. Although a number of studies
have documented and some attempted to explain this stylized fact, the
issue still remains unresolved. In a multi-period bargaining framework
they show that if firm-specific training is more productive in foreign
firms, foreign firm workers will have a steeper wage profile and thus
acquire a premium over time. Using a rich employer-employee matched data
set for Ghanian manufacturing they show that the foreign wage premium is
only acquired by workers over time spent in the firm and only by those
that receive on the job training, thus providing empirical support for a
firm-specific human capital acquisition explanation.
Karolina
Ekholm (NHH/SNF, LSE and SSE) and Katariina Hakkala (SSE) have developed a
theoretical model where firms choose the location of R&D activities
and production activities separately. In deciding where to locate
production, proximity to large consumer markets becomes important for
intermediate levels of trade costs. In deciding where to locate R&D
labs, it is assumed that the firms take into account the potential
benefits stemming from knowledge spillovers created by other firms’
R&D activities in the same region. The model intends to capture the
main features of many high-tech industries, where knowledge spillovers are
important and where firms operating in the industries typically are
multinationals with activities in several countries. The analysis shows
that for a range of trade costs where agglomeration economies in final
goods production are especially strong, a small country may become
specialised in R&D activities. Both production of high-tech goods and
R&D activities are assumed to require inputs of skilled labour. When
agglomeration tendencies are particularly strong, the price of skilled
labour becomes so high in the large country that there are cost advantages
from carrying out R&D activities in a small country. The analysis also
shows that for a large set of parameter values there are multiple
equilibria with R&D becoming concentrated in either the small or the
large country.
At
UPF the main focus of theoretical modelling has been on the integration of
the classical theory of the multinational firm and the more recent theory
of the firm based on incomplete contracting. Several papers have been
produced relating to these issues. The first, ‘Foreign Direct
Investments and Spillovers through Workers’ Mobility’, by Thomas Rønde
(UPF), Massimo Motta (EUI and UPF) and Andrea Fosfuri (Universidad
Carlos III, Madrid) focuses on the problem of spillovers through movement
of personnel from the multinational to firms in the local economy.
Multinationals can avoid these information transfers occurring by
resorting to exports. The paper establishes conditions under which these
type of information transfers will occur in equilibrium. Motta and Ronde
also collaborated on work on ‘Trade Secret Laws, Labour Mobility and
Innovations’. The resulting paper shows that when the researcher’s
(observable but not contractible) contribution to innovation is crucial, a
covenant not to compete (CNC) reduces effort under both spot and
relational contracts. Having no CNC allows the researcher to leave for a
rival. This alleviates the commitment problem by allowing the firm to
reward a successful researcher. However, if the firm’s R&D
investment mainly matters then a CNC is optimal, as it ensures the
firm’s incentives to invest. Finally, strong trade secret laws are
optimal if relational contracts are sustainable, but not if spot contracts
are used. This is particularly important for multinational firms investing
in foreign countries, as spillovers from their R&D effort into foreign
countries where their knowledge capital is less protected than at home.
As
part of the project, Antonio Cabrales (UPF) has been researching pharmaceutical
generics, vertical product differentiation, and public policy. The
resulting paper studies oligopolistic competition in off-patent
pharmaceutical markets using a vertical product differentiation model.
This model can explain the observation that countries with stronger
regulations have smaller generic market shares. It can also explain the
differences in observed regulatory regimes. Stronger regulation may be due
to a higher proportion of production that is done by foreign multinational
firms. Finally, a closely related model can account for the observed
increase in prices by patent owners after entry of generic producers.
In
‘Multinational Firms and Quality Competition’, Stephen Pavelin (UCD)
and Steve Brammer explored the impact of quality competition on the
pattern of horizontal FDI across countries with asymmetric market sizes.
The effect of host market size on horizontal FDI flows has been addressed
by standard explanations of MNE activity – concluding that larger
markets are more attractive destinations for FDI. This predicts large
flows of FDI into countries with large markets, from wherever. Pavelin and
Brammer argue that what matters is not only the size of the host country
market, but also the relative sizes of the home and host country markets.
This is because of the effect that quality competition in firms’
domestic market may have on the generation of firm-specific assets. In
this framework, the incentive for FDI is shown to be greatest when there
is some intermediate level of asymmetry in home and host country size –
with FDI flowing from the larger country, to the smaller country.
A
second paper of Pavelin’s, co-authored with Dermot Leahy (UCD), is
entitled ‘Foreign Production and Collusion: Knickerbocker Revisited’.
In this paper, the interdependence of firms’ FDI decisions as proposed
by Knickerbocker in the early 1970s is revisited and formalised. However,
by framing FDI as that accompanying vertically-related multinational
production and concentrating on the sustainability of tacit collusion, the
paper diverges from the specifics, but not the spirit, of
Knickerbocker’s work.
In
‘Follow-my-leader FDI and Collusion’, Leahy and Pavelin (both UCD)
present a simple model to express the idea that domestic rivals may be
motivated to set up foreign production in the same country because the
replication of each other’s FDI facilitates collusive behaviour in the
market in which they compete. This
implies positive inter-dependence between firms’ FDI decisions; i.e.
foreign investment by one firm offers increased incentives for others to
follow suit. Thus the paper
highlights a mechanism that propagates FDI clusters. In a paper entitled
‘Strategic Interaction and the Domestic Market Effect: A Model of
the Choice Between Exporting and Multinational Production with
Cost-Reducing R&D Expenditures’, Pavelin (UCD) also developed a
model to illustrate the interaction between two firms based in different
countries, each of which faces the export vs MNE choice concerning the
servicing each other’s home market. The basic game structure is similar
to that elsewhere in the literature. To this, a further choice is added.
Here firms not only interact in their export vs MNE and output choices,
but also in their choice over investment in a new technology that allows a
corporate-wide reduction in variable costs (i.e. cost-reducing R&D).
In the presence of such corporate-wide investment, firms’ decisions
concerning each other’s home markets are interdependent. Furthermore,
the strategic motives for FDI relate not only to a firm’s foreign market
profits, but also to those from their domestic market. This is because one
firm’s export vs MNE choice can influence both its rival’s choice and
investment behaviour. One possibility is that a firm sets up a plant
overseas in order to influence the behaviour of its rival, even though its
profits from serving the foreign market would be higher by exporting.
In
‘Divide and Rule: Geographical Diversification and the Multinational
Firm’, Dermot Leahy and Stephen Pavelin (both UCD) present a model where
a firm chooses how many plants to have.
It is shown that a firm that initially has a plant in its home
country may choose to have a foreign plant to improve its bargaining
position versus local labour unions. This permits the firm to secure lower
wages than if they remained domestic. Furthermore, choosing to have a
plant in more than one foreign country may lower wages further. Thus, the
firm is faced with a potential link between the wage rate and its degree
of geographical diversification.
As
indicated by the publications cited in the references to this report,
Carsten Eckel, of the NHH/SNF team, has worked on a variety of theoretical
papers related to FDI, international fragmentation of production,
employment and relative wages. Also
from NHH/SNF, Karolina Ekholm (SSE, NHH/SNF and LSE) and Karen-Helene
Midelfart Knarvik (NHH/SNF)
have explored how market
integration may induce a skill-biased technology change as firms are
induced to shift to technologies implying higher fixed costs and lower
variable costs. Moreover, this may explain how increased integration has
led to a general increase in firm size – reflected through the rise in
number of multinational companies and cross-border mergers and
acquisitions. The authors analysed the implications of such a technology
change for skill premiums and skill intensity.
Henrik
Braconier (IUI, Stockholm), Pehr-Johan Norbaek (IUI, Stockholm) and Dieter
Urban (CSLA and LSE) have worked on a new empirical functional form for
the knowledge capital model that is more directly attached to the theory
than the one of previous studies. They have collected a new dataset with
the largest possible coverage of cross-country data on FDI stocks and
affiliate sales from private and public national and international
sources. Strong evidence for the knowledge capital model is found,
regardless of the skill measures used. They have also shown that the
specification previously used by Carr, Markusen and Maskus implies an
inexact measure of skill abundance, which in addition to insufficient data
coverage, may explain why the vertical part of the model is not manifested
in previous studies.
Gilles
Duranton, of the LSE team, has been working on a paper which introduces
the concept of production systems. He assumes a standard thick-market
externality together with the fact that higher-quality goods also require
higher skills from workers. Firms face a trade-off between low-quality
goods with low skill requirements for which the potentially abundant
labour force generates strong thick-market externalities and
higher-quality goods with higher skill requirements. In equilibrium, the
economy is partitioned into production systems, i.e. clusters of firms
producing the same quality. The distribution of skills determines the
boundaries of the production systems, which in turn determine the wages.
In this framework an increase in the supply of skilled workers can induce
first higher wages for all workers and then higher wages for the skilled
but lower wages for the unskilled. This is consistent with the late 20th
century evolution of the US labour market.
This
time with Diego Puga (LSE), Gilles Duranton has also worked on a paper
which develops a model of individual firms’ decisions on whether or not
to fragment their management and production activities to the aggregate
production structure of different locations. With Dan Trefler this time,
Puga has also worked on a positive theory of the incentives a firm uses to
induce incremental innovation on the part of its employees and
subcontractors. The objective is to understand the role of incremental
innovation for the internal organization of the firm.
Determinants
of the location of FDI projects
The
NHH/SNF team has worked on models along the lines of the ‘new economic
geography’, focusing on the location of production and driving forces of
locational choice. To combine insights from the traditional literature on
FDI and MNCs with the recent literature on the new economic geography, Jan
Haaland (NHH/SNF) and Ian Wooton (University of Glasgow and CEPR) (1998)
have explored the impact of multinational corporations on host economies
when technological and pecuniary externalities are important determinants
of firms’ choice of location. Haaland and Wooton have also looked
specifically at the role of labour market conditions for the location of
FDI. In their 2002 paper they explore the impact of various labour market
characteristics on the establishment of firms operating in widely
differing business climates in terms of degree of risk. They find that in
a game between two countries, a nation with an inflexible labour market
and high unemployment will succeed in attracting low-risk firms, while one
with more flexible labour markets and low unemployment will win the game
for higher-risk firms.
The
interaction between capital mobility, FDI, taxation and economic
integration are issues that have been explored by Hans Jarle Kind,
Karen-Helene Midelfart Knarvik and Guttorm Schjelderup (all NHH/SNF). In
particular, their studies provide insights into how the presence of
agglomeration forces impact on optimal taxation. The work has resulted in
a number of papers as referenced. Similar topics were covered by Forslid,
one of the NHH/SNF young researchers, in Andersson and Forslid (1999).
Kind, Midelfart Knarvik and Schjelderup have also explored the interaction
between different corporate tax regimes (separate accounting and formula
apportionment) and increased integration. The question they address (see
Kind, Midelfart Knarvik and Schjelderup, 2000) is whether the increased
number of multinationals which actively use transfer pricing as a tool for
profit sharing, and falling trade barriers call for a change in tax regime
at the European level.
Also
working in the field of the ‘new economic geography’, Mori and Turrini,
of the CSLA team, developed a model investigating the role of skill
heterogeneity in explaining locational patters induced by pecuniary
externalities. A symmetry-breaking result is obtained: symmetric
configurations cannot be stable, and regional inequality is inevitable.
The relatively more skilled choose to stay in the location with higher
aggregate income and skill, while the relatively less skilled stay in the
other. The model allows the authors to analyse the links between the
extent of interregional inequality and the extent of interpersonal skill
inequality.
Anthony
Venables (LSE) has studied the geography of investment to show that
international capital flows will typically go to geographically-favoured
regions – despite higher wages in these regions – tending to leave
less-favoured regions at best unaffected, and at worst damaged. A
socially-optimal allocation of capital would typically place more capital
in disadvantaged regions than does the market equilibrium.
Enrico
Pennings, of the UPF team, prepared a paper on the ‘Choice and Timing of
Foreign Market Entry under Uncertainty’ (with Leo Sleuwaegen). The work sheds
new light on why timing and entry mode should be considered
simultaneously. They derive the profit levels at which it is optimal to
switch from exporting to setting up a wholly-owned subsidiary, creating a
joint venture, or licensing production to a local firm. The preferred
entry mode depends on uncertainty about future profits, tax differentials
between the home and the foreign country, the cost advantages of local
firms, institutional requirements, and the degree of cooperation between
partners in a joint venture.
A
second paper by Pennings, ‘How to Maximize Domestic Benefits from
Irreversible Foreign Investments’, assumes a foreign monopolist who can
either export to a host country, or undertake an irreversible FDI. It is
shown that the host government maximizes net domestic benefits by nearly
fully subsidising the investment cost in combination with taxing away
benefits that exceed the gains from exporting. Since a higher tariff
increases the firm's propensity to invest and increases tax benefits,
maximizing net domestic benefits yields an optimal tariff that is higher
than the one derived in previous studies that disregard the dynamics of
FDI and the interaction between optimal tax and tariff policy.
In
‘Foreign Investment and the Single Market’ Peter Neary (UCD) extends
the theory of MNCs to explore the effects of internal trade liberalisation
by a group of countries on the level of inward direct investment. The
analysis identifies three distinct influences on how an MNC chooses to
serve the union markets. First is the tariff-jumping motive. Less familiar
from the existing literature is the idea that reductions in internal
tariffs reduce the tariff-jumping incentive to establish more than one
union plant, and this encourages plant consolidation. Second is the export
platform motive. As internal tariffs fall, FDI with only a single union
plant is favoured relative to exporting. Finally, reduced internal tariffs
lead to increased competition between domestic firms, which dilutes both
the tariff-jumping and export-platform motives. This works against both
FDI and exports and may lead to the ‘Fortress Europe’ outcome of
multinationals leaving union markets even though external tariffs are
unchanged.
In
‘EU Accession and Prospective
FDI Flows to CEE’ Frank Barry (UCD)
points out that most
current CEE-bound FDI comes from Europe rather than the US, is market
seeking rather than export oriented, and is relatively low-tech. The same
situation prevailed in Ireland before that country joined the EU, but the
situation was rapidly reversed upon accession.
This paper suggests that uncertainty about CEE public policy, CEE
public administration and the timing of CEE accession explains why
high-tech export-oriented multinationals have not yet begun to invest
heavily in the region.
The
impact of foreign direct investment on host economies
In
his August 2000 paper, Kristof Dascher (UCD) has developed a model of an
Irish-type FDI-driven boom that generates agglomerations without recourse
to the assumption of technological externalities. This model of a ‘very
open’ small economy, in which labour and capital as well as goods are
internationally mobile, incorporates labour inflows that reinforce the
capital inflows which initially cause the boom. The non-traded fixed
factor which ties down the equilibrium is land; the economic boom is
therefore associated with congestion. Using the tools of duality theory,
Dascher analyses the distribution of welfare gains and losses over the
various groups of factor owners in the economy. His 2001 paper applies
aspects of this model to the Irish situation.
This work is developed further in Dascher’s 2002 paper which points out
that, as in the previous research, FDI creates gains for some residents
and losses for others. To win support for FDI local governments may want
to pay cash compensation. In the model cash payments are unsuccessful,
however, although public housing is. Dascher argues that public housing
makes FDI more acceptable where cash transfers fail, so local governments
may choose to invest in public housing to overcome opposition to FDI. The
paper presents supporting case studies from Hong Kong and Singapore.
Regional
issues and congestion are also taken up by Frank Barry in his paper on
‘FDI, Infrastructure and the Welfare Effects of Labour
Migration’. The paper
develops a model of a small open economy with open capital and labour
markets, where labour demand is based on capital mobility and increasing
returns in production. Migration decisions are based on the relative
attractiveness of regions in terms of the stock of infrastructure,
including its tax cost, the degree of congestion, and the prevailing level
of wages. Equilibria are not Pareto-efficient because individuals do not
take account of the impact of their actions on the level of wages
prevailing, the extent of the tax base to finance infrastructural
provision, or the degree of congestion. The model generates new insights
into a range of policy issues that surfaced over the course of the recent
Irish boom.
Dermot
Leahy (UCD) and Catia Montagna (University of Dundee) have explored, at
the theoretical level, the relationship between labour-market unionisation
and FDI. ‘Unionisation and FDI: Challenging Conventional Wisdom’ shows
that labour market unionisation can cast doubt on some aspects of the
conventional wisdom on FDI. In particular, they show that it is not always
welfare-improving to attract inward FDI, and that MNEs may prefer
centralised to decentralised wage-setting regimes.
Dieter
Urban, of the CSLA team, has analysed the issue of employment risk that
may accompany the arrival of MNCs. He developed a model that compares, in
a simple general equilibrium setting, a regime with national firms to a
regime with multinational firms in the presence of consumption risk
insurance by implicit contracts. The model shows that terms-of-trade
behaviour is different with insurance. An optimal policy rule is derived
that imitates the optimal insurance equilibrium if there is a lack of
private contract enforcement. Employment, real wage, and profit
fluctuations are compared in the two regimes in response to technology
shocks.
With
Rachel Griffiths and Helen Simpson, Stephen Redding of the LSE team has
worked on a paper which investigates whether there is convergence in total
factor productivity at the establishment level, to the technological
frontier. The authors provide evidence of convergence to the frontier,
suggesting the existence of technology spillovers. Foreign multinationals
make up a significant proportion of establishments at the technological
frontier. This implies that high productivity firms, both domestic and
foreign-owned, make a contribution to productivity growth through
technology transfer. They also find evidence that increased foreign
presence within an industry raises the speed of convergence to the
technological frontier.
Marcus
Haacker (LSE) has undertaken research on the role of multinationals in the
dissemination of new technologies and know-how; analysing, for example,
the interactions between multinationals and the host economy through the
labour market.
Policy
towards FDI
In
‘Temporary Social Dumping, Union Legalisation and FDI’, Dermot Leahy (UCD)
and Catia Montagna (University of Dundee) show that a developing
host-country government may have an incentive to adopt temporary social
dumping, in the sense of preventing unionisation in the short run in order
to attract FDI, which allows higher rents to be extracted in the future.
If the government has recourse to a fiscal instrument in conjunction with
union legalisation, however, the need to engage in social dumping can be
circumvented.
Alessandro
Turrini and Dieter Urban (both CLSLA), working on policies towards FDI,
provided a theoretical underpinning to the position of many developing
countries (LDCs) which are against the implementation of a Multilateral
Agreement on Investment (MAI). In their model, participation in MAI
involves a trade-off between less rent extraction from MNEs and more
abundant FDI inflows. At equilibrium, either all countries enter MAI, or
all countries stay out, or only some of them enter. Coordination problems
may induce multiple equilibria, which may co-exist. So, the implementation
of an MAI may depend not only on structural factors, but also on the
general ‘political climate’. When all countries join MAI, world
welfare is maximized because this minimizes the hold-up problem faced by
MNEs and stimulates investment. However, in an asymmetric world, welfare
gains for all countries are not guaranteed.
2. Econometric Analysis
As
an initial step towards the econometric analysis to be undertaken, the
CSLA team prepared a paper on available data sources on FDI and
multinational companies. The paper reviewed sources, characteristics,
availability and limitations of databases on FDI and MNCs, both at a
national and international level. The survey was intended to provide a
guide to FDI and multinationals data for empirical work, examining three
different types and sources of information. First, it examines
international guidelines for the compilation of balance of payments and
direct investment position data. Second, it reviews the main
characteristics of international statistics on FDI flows and stocks,
considering the major available international sources (OECD, EUROSTAT, UN,
etc.). Finally, it illustrates financial and operating data on MNCs
compiled by individual countries.
A
database was constructed by Dieter Urban (CSLA and LSE) with bilateral FDI-stock
data and bilateral affiliate sales data from private sources, OECD sources
(Globalization database) and from national statistics of US, Swedish,
Japanese, German and Italian outward FDI data as well as inward FDI data
of about a dozen countries for the years 1986, 1990, 1994, and 1998. In
total, about 60 home and 60 host countries are contained with at least 1
observation. Both FDI stock and affiliate sales data exhibit large
data-errors, when comparing the same observations from inward and outward
FDI sources. The correlation between FDI stocks, which is the traditional
measure of FDI activity, with the newly assembled data on affiliate sales
is a mere 0.79. Hence, studies which proxy affiliate activity by FDI
stocks rather than affiliate sales must be considered with caution.
Karolina
Ekholm (NHH/SNF, SSE and LSE), Karen-Helene Midelfart Knarvik (NHH/SNF),
and Henrik Braconier (IUI), have looked at the role of MNCs in
transmitting technology across national borders. Using Swedish
industry-level data, their paper analyses whether inward and outward FDI
work as channels for international R&D spillovers. They find no
evidence of FDI-related R&D spillovers; at neither the firm nor the
industry level in Swedish manufacturing. The only variable that
consistently affects total factor productivity is own investment in
R&D.
One
of the important questions that this project seeks to answer is what
drives the choice between FDI and alternatives such as licensing, joint
ventures or export. Research undertaken by the UPF team analysed this
organizational decision process for the chemical industry. ‘The
Organization of Production in the Chemical Industry’, by Bruno Cassiman
(UPF), Alfonso Gambardella (University of Urbino) and Walter García
Fontes (UPF), for example, involved the construction of a comprehensive
data set for FDI, mergers and acquisitions, joint ventures and licensing
behaviour in the chemical sector between 1985 and 1997.
In
‘Firm Interdependence in
Foreign Production: Leading UK Firms in 1986 and 1993’, Stephen Pavelin
(UCD) estimated econometric models explaining the foreign production of
leading UK firms in 1986 and 1993. The paper employs firm-level data
describing the world-wide production of each of these firms, disaggregated
by industry and geographical region. The principle questions addressed
are: (i) What effect does one UK firm’s foreign operations have on the
foreign operations of another UK firm? and (ii) What effect does the UK
operations of a foreign firm have on the foreign operations of UK firms?
Pavelin finds strong evidence of a negative interdependence between the
foreign operations of UK firms.
In
‘Multinational Enterprises and New Trade Theory: Evidence for the
Convergence Hypothesis’, Salvador Barrios (UCD), Holger Görg
(University of Ulster, SSE and University of Nottingham) and Eric Strobl (UCD)
analyse the following issue. According to the ‘convergence hypothesis’
MNCs will tend to displace national firms and trade as total market size
increases and as countries converge in relative size, factor endowments,
and production costs. Using a recent model developed by Markusen and
Venables (1998) as a theoretical framework, the paper explicitly develops,
and address the properties of empirical measures to proxy displacement of
national by multinational firms between two countries. These empirical
measures are then used to test the convergence hypothesis for a panel of
data of country pairs over the years 1985–96. The results provide some
empirical support for the convergence hypothesis.
Barry,
Görg and Strobl explore empirically the distinction between ‘efficiency
agglomerations’ and ‘demonstration effects’, both of which can lead
to a cascade of FDI. They find evidence of both at work in the Irish
economy. (Full details are given in the Irish Economy Case Study below).
In
‘Multinational Companies and Productivity Spillovers: A
Meta-Analysis’, Holger Görg and Eric Strobl present the results of a
meta-analysis of the literature on multinational companies and
productivity spillovers. Studies in this literature examine spillovers
usually within the framework of an econometric analysis in which labour
productivity in domestic firms is regressed on a number of covariates
assumed to have an effect on productivity, one of which is the presence of
foreign firms. A positive and statistically-significant coefficient on the
foreign presence variable is then taken as evidence that spillovers exist.
For a sample of published and unpublished studies, the different
coefficients on the foreign presence variable reported in different
studies, and their associated values of the t-statistics, are collected.
The values of the t-statistics are then regressed on a number of study
characteristics, such as sample size, variable definitions used, etc..
Some of these characteristics, namely, variable definitions, and whether
it is a cross-section or panel analysis, have an effect on the size of the
coefficient found in the productivity studies. Using a similar regression
approach, the study also finds evidence that there may be publication bias
in the literature on productivity spillovers.
Over
the course of the project Frank Barry and Aoife Hannan (both UCD) also
produced two empirical papers of more than just Irish interest.
‘Product Characteristics and the Growth of FDI’ analyses the
growth in the world FDI-to-GDP ratio over recent decades. Previous
explanations posited a growth in FDI relative to value-added within
sub-sectors, either because of capital-market liberalisation or changes in
market size. This paper focuses on the relative growth of sub-sectors with
high FDI to value-added ratios, showing that these sub-sectors have higher
income elasticities of demand than is the case for aggregate manufacturing
or services. The policy implications of these findings are also explored.
Since
first proposed by Balassa, indicators of revealed comparative advantage
(RCA) derived from current production and trading patterns have been used
frequently to predict the sectoral effects of trade liberalisation. The
paper ‘FDI and the Predictive Powers of Revealed Comparative Advantage
Indicators’, by Barry and Hannan, identifies a serious flaw in the
methodology. The paper shows that it would have failed completely to
predict post-EU-accession changes in Ireland’s sectoral structure and in
sectoral export performance. These developments were instead driven by the
country’s success in attracting FDI, and the sectoral destinations of
these greenfield FDI inflows were unrelated to measures of the country’s
pre-accession RCA. It goes on to show, however, that the methodology is
reasonably accurate as a predictor of developments in indigenous (i.e.
domestically-owned) industry in Ireland. The conclusion is that measures
of revealed comparative advantage will be inaccurate predictors of
structural change in countries which are successful in attracting
substantial inflows of greenfield FDI. The most advanced CEE countries,
for example, display characteristics that may be indicative of future
success in this regard.
At
LSE, Anthony Venables used trade data to address the issue of production
networks. The dataset provides bilateral trade flows in parts and
components, together with trade in the associated final product, at a fine
level of commodity disaggregation. It is possible to identify sectors in
which either upstream or downstream stages of production are being
undertaken in lower-wage countries and investigate the growth of these
outsourcing activities through time. The world-wide pattern of FDI was
reviewed by Howard Shatz and Tony Venables. They show that, while the
overwhelming proportion of FDI is horizontal in nature, a growing
proportion, particularly to developing countries, is now vertical.
Effects
of FDI on the home economy
In
this section of the project the CSLA team has concentrated its attention
on the home labour market effects of international production by
multinational firms. The issue of whether employment abroad complements or
substitutes employment in parent companies was investigated by Bruno and
Falzoni. Using industry-level data on US MNCs for the period 1982-1994,
they test for the presence of labour adjustment costs and estimate
short-run and long-run price elasticities for labour demands in different
locations. The findings show that, due to slow input adjustments, the
complementarity/substitution relationship between employment in US parents
and employment in Latin American affiliates is reversed from the short to
the long run. While in the short run there is evidence of labour
substitution, in the long run a complementarity relationship emerges,
suggesting a vertical division of activities. Differently, labour
substitution seems to prevail both in the short run and in the long run
between affiliates' locations in the Western Hemisphere (North and Latin
America) and in Europe.
Also
under this topic Barba Navaretti, Bruno, Castellani and Falzoni and
Barba Navaretti and Castellani investigated the employment effects
of FDI on parent companies in Italy. The two studies use a new data set
combining data on the parent company and on the subsidiaries of Italian
multinationals. In addition, investing firms are compared to a
counterfactual of firms that have not invested abroad. Adopting different
empirical methods, both studies find support of a positive effect of
investing abroad on performance at home.
Another
issue analysed by the CSLA team was the effect of the greater ease in
relocating production on labour demand elasticity. When economies become
more integrated, competition in product markets will increase and the
demand for labour will generally become more elastic. Moreover, both trade
and the enhanced international mobility of firms will make domestic labour
more substitutable with foreign factors of production. The market power of
unions will thus decline. This issue is particularly important in Italy,
as well as in other European countries, because of the prominent role of
trade unions in the wage setting process. Using Italian data, some support
is found for the hypothesis that greater globalisation is associated with
larger elasticities. The sectors with a higher share of employees in
foreign affiliates, a proxy of the level of multinational involvement, or
with a higher degree of trade openness show a high elasticity of labour
demand (Faini et al.). In a different study, using data from a number of
industrialised countries, including major European countries and the US,
Bruno, Helg and Falzoni find mixed results on the hypothesis of increasing
labour demand elasticity.
The
impact of the activity of investing abroad on parent firms’ productivity
was investigated by Castellani using a panel of Italian firms. Following
studies that address a similar question with reference to exports, the
learning effects from outward investments were measured by looking at the
stochastic process governing productivity growth. The
learning-by-investing hypothesis is tested specifying a partial adjustment
process for productivity, where foreign investments enter as a
predetermined regressor. The results show a drop in productivity growth in
the year of investment, followed by a rise in later years, which outweighs
the short-term fall.
Barba
Navaretti, Castellani and Zanfei addressed the same issue using two novel
firm-level datasets on Italy, France and Spain. The empirical analysis
shows that firms investing abroad between 1993 and 1997 have improved
their competitiveness and efficiency.
Bruno
Cassiman (UPF) and Reinhilde Veugelers (Catholic University of Leuven and
CEPR) have worked on the issue of international information flows between
countries and multinationals. The paper ‘Importance of International
Linkages for Local Know-How Flows: Some Econometric Evidence From
Belgium’ assesses econometric
evidence specifically from Belgium. External knowledge is an important
input for the innovation process of firms. Increasingly, this knowledge is
likely to originate from outside of their national borders. This explains
the preoccupation of policy-makers with stimulating local technology
transfers coming from international firms. They find that firms that have
access to the international technology market are more likely to transfer
technology to the local economy. In doing so, they qualify the traditional
assertion that multinational firms are more likely to transfer technology
to the local economy. Once controlled for the superior access to the
international technology market that multinationals enjoy, they find that
these firms are not more likely to transfer technology to the local
economy compared to exporting or local firms that have access to the
international technology market.
Another
paper by Cassiman and Veugelers, entitled ‘Innovative Strategies and
Know-how Flows in International Companies: Some Evidence from Belgian
Manufacturing’ tries to empirically assess how technology flows are
structured in international firms. While all types of international firms,
including subsidiaries, are found to be more innovation active than local
firms, companies which are part of an international group, as affiliates
but especially as headquarters, have the widest innovation strategy,
relying on internal as well as external sources. These external sources
are located nationally as well as internationally, and are accessed
through buying and cooperative strategies. In addition, internal transfers
and intra-group cooperation are quite pervasive in these companies,
although the evidence for transfers from headquarters to subsidiaries is
stronger than for the reverse flow from subsidiaries to headquarters.
Chiara
Fumagalli (Università Bocconi, Milano) and Massimo Motta (EUI, Florence,
and UPF) wrote the paper ‘On the Relocation of Economic Activities’,
which analyses the concern that globalisation might lead to a considerable
relocation of production and employment out of the developed countries.
The question is whether this concern is sound and, more generally, whether
FDI (relocation is a particular type of FDI) really has such negative
effects on the home economies as critics claim. The authors find that
theoretical arguments predict ambiguous effects of relocation on the
country of origin, so that whether such effects are positive or negative
is mostly an empirical question. Surveying the empirical evidence, they
find that there exists no convincing evidence that outward FDI and
relocation of operations are actually harmful on average. Rather, it seems
that in many cases there exist complementarities between foreign and home
production, so that investing abroad might actually increase domestic
production.
The
SSE team has conducted several econometric studies on the issue of how the
activities by multinational firms affect the demand for labour in their
home countries. It has been argued that because many multinationals are
footloose in the sense that they can easily relocate activities from one
location to another, they contribute to increased wage competition between
countries. This presupposes that the multinationals respond to changes in
relative wage costs by relocating activities from high-cost locations to
low-cost locations. The SSE team has carried out a number of studies
trying to quantify the extent to which such relocation of activity takes
place. One result found by Henrik Braconier (IUI, Stockholm) and Karolina
Ekholm (NHH/SNF, LSE and SSE) in a study based on Swedish data is that
whereas such relocation seems to take place at the level where firms
decide whether to set up production in a particular country or not, there
is no evidence of this taking place within existing production units. In
fact, Braconier and Ekholm find very small effect of changes in relative
wage costs on employment within existing production units.
Magnus
Blomström (SSE), Eric Ramstetter (ICSEAS) and Robert Lipsey (NBER) have
carried out work on US multinationals which appear to locate
labour-intensive activities abroad, thereby reducing demand for labour in
the home country. They do not find a similar pattern for Japanese and
Swedish firms, however. For the latter, supervisory and ancillary
employment at home to service foreign operations seem to outweigh any
relocation of labour-intensive production so that the net effect of a
foreign expansion is to increase employment in the parent firm.
Effects
of FDI on the host country
Koen
de Backer (UPF and KULeuven) has worked on ‘Does FDI crowd out Domestic
Entrepreneurship?’. In this paper he analyses firm entry and exit across
Belgian manufacturing industries and presents evidence that import
competition and FDI discourage entry and stimulate exit of domestic
entrepreneurs. These results are in line with theoretical
occupational-choice models which predict that FDI would crowd out domestic
entrepreneurs through their selections in product and labour markets.
However, the empirical results also suggest that this crowding-out effect
may be moderated or even reversed in the long run due to the long-term
positive effects of FDI on domestic entrepreneurship as a result of
learning, demonstration, networking and linkage effects between foreign
and domestic firms.
In
another line of research, de Backer studies ‘Productivity Dynamics in
Foreign-owned Firms’. He looks at the distinctive contribution of
foreign subsidiaries and domestic firms to productivity growth in
aggregate Belgian manufacturing to show that foreign ownership is an
important source of firm heterogeneity affecting productivity dynamics.
Foreign firms have contributed a disproportionate share of aggregate
productivity growth. More importantly, reallocation processes differ
significantly between the groups of foreign subsidiaries and domestic
firms.
In
their paper ‘On the Determinants of Multinationals’ Ownership
Preferences’ Natalia Barbosa and Helen Louri of the IMOP team addressed
the issue of the ownership structure that MNCs select when investing
abroad. This is an important question for both the MNC and the domestic
partner as it affects the profitability of invested assets. Furthermore,
the degree of foreign involvement may affect (through spillovers) the
general performance of the host economy. The evidence from Greece and
Portugal displays different ownership preferences despite the similarities
of the two countries. Using data from these two countries they find that
both firm and industry characteristics interacting with location affect
ownership decisions. A more recent collaboration between Barbosa and Louri
resulted in the paper ‘Corporate Performance: Does Ownership Matter?’
in which the authors investigate whether MNCs operating in Portugal and
Greece perform differently than domestically-owned firms. They find that
ownership ties do not make a significant different with respect to
performance of firms operating in Portugal, but that MNCs operating in the
Greek market are significantly more profitable than Greek-owned firms.
Also
from IMOP, Sophia Dimelis, with Helen Louri, analysed the production
efficiency gains in terms of technology transfer and labour productivity
caused by diverse degrees of foreign ownership using a sample of 4056
manufacturing firms operating in Greece in 1997. The work, published as
‘Foreign Ownership and Production Efficiency: A Quantile Regression
Analysis’, finds a positive effect on labour productivity of foreign
ownership, which stems exclusively from fullly- and majority-owned
affiliates and becomes significant only in the middle quantiles.
Productivity spillovers benefiting local firms are also differentiated,
with minority holdings exercising a stronger effect in most quantiles.
Subsequent work by these two authors addressed further the efficiency
benefits of FDI to host economies by analysing whether they were produced
in equal measure by all foreign firms, and the extent to which such
benefits were distributed evenly across domestic firms. The results
indicate that foreign firms are more productive than domestic firms, and
this difference increases the higher the foreign ownership share. When
spillovers are taken into account, while a general positive net effect is
expected, it becomes evident that significant positive spillovers stem
only from firms with minority foreign ownership and are enjoyed
exclusively by small firms.
In
’Foreign Investment and Ownership Structure: An Empirical Analysis’,
Helen Louri, Raymond Loufir and Marina Papanastassiou (all IMOP) examined
the micro-determinants of the degree of ownership MNCs select when
expanding their production abroad. Using data on 216 foreign firms located
in Greece in 1998, they find that firm and industry characteristics,
through their effects on expected returns, shape ownership decisions.
Linked to this work is Louri’s research with E Dedousis, published as
‘Corporate Governance and Investment: Domestic and Foreign Firms in
Greece’. Here the authors test the hypothesis that ownership affects
investment on its own and interacting with return expectations and cash
flow.
This
time with Georgios Fotopoulos (University of Thessaly), Helen Louri also
carried out research to enhance understanding of the empirical
determinants of corporate growth. ‘Corporate Growth and FDI’ aimed to
extend the literature to include a new group of variables related to FDI,
namely the degree of foreign ownership and technology spillovers. The
analysis also takes into account the role of sunk costs and financial
structure. They find that the role played by MNCs in increasing corporate
growth varies in intensity depending on industry groups and also that the
use of new variables is justified.
Whether
the presence of MNEs benefits local economies by promoting learning and
catch-up of local firms was investigated by Giovanni Peri and Dieter Urban
of CSLA. Such a channel of spillovers from MNEs to local firms is known as
the Veblen-Geschenkron effect. Rather than the overall density of MNEs in
a region or sector, it is their initial productivity advantage on the
local firm to determine the positive effect on domestic productivity
growth. They test this hypothesis using firm-level data for German and
Italian companies during the 1990s and they find evidence of a significant
and robust Veblen-Gerschenkron effect.
Using
a large firm-level data set covering all sectors of Spanish manufacturing
during the period 1983-1996, Alessandro Sembenelli and Georges Siotis
(both CSLA) attempted to disentangle two expected effects of the presence
of MNCs in an economy: a) an increase in average productivity following a
wave of FDI as MNCs enjoy higher levels of efficiency; and b) an increase
in competitive pressure on the domestic market. By estimating a dynamic
model of firm-level profitability, they find that FDI has a positive
long-run effect on the profitability of target firms, but this is limited
to firms belonging to R&D-intensive sectors. In addition, the results
indicate that foreign presence dampens margins. However, this effect
appears to be more than compensated by positive spillovers in the case of
knowledge intensive industries.
Analysing
the effect of foreign ownership on labour demand, Giorgio Barba Navaretti,
Checchi and Alessandro Turrini
(all CSLA) provided a cross-country study based on firm-level data. They
estimated labour demand equations in eleven European countries using
dynamic panel data techniques on samples that permit to distinguish the
ownership status of firms. They find that the employment adjustment is
significantly faster in MNCs' affiliates, irrespective of the country
investigated. As for the wage elasticity of labour demand, MNCs show
smaller elasticities compared with national firms, and very little
variation across countries. Cross-country correlations show that the
relative value of wage elasticities in MNCs on that in national companies
is positively related to country-level indexes of labour market regulation
(employment protection, union presence,...). MNCs seem to have a more
rigid demand for total labour (possibly due to a different skill
composition). However, being MNCs relatively ‘footloose’, this
difference tends to vanish as the rigidity of employment regulations
rises.
Holger
Görg (University of Ulster, University of Nottingham and SSE) and Erik
Strobl (UCD) have conducted an empirical investigation of spillovers from
foreign firms through worker mobility. While there has been a large
empirical literature on productivity spillovers from foreign to domestic
firms this literature treats the channels through which these spillover
effects work as a black box. The work of Görg and Strobl attempts to fill
this gap in the literature by examining spillovers generated through
labour mobility in Ghanian manufacturing. The results suggest that firms
which are run by owners that worked for MNCs in the same industry
immediately prior to opening up their own firm have higher productivity
growth than other domestic firms. This suggests that these entrepreneurs
bring with them some of the knowledge accumulated in the MNC which can
usefully be employed in the domestic firm. They do not find any positive
effects on firm-level productivity if the owner had experience in MNCs in
other industries, or received training by MNCs.
Fredrik
Sjöholm (SSE) and Robert Lipsey (NBER) have carried out a study of the
effect of inward FDI on the labour market of the host country using data
from Indonesia. They find that higher foreign presence in an industry
and/or region is associated with higher wages for two reasons: (i)
foreign-owned firms tend to pay higher wages for a given job, and (ii)
higher foreign presence tends to lead locally-owned firms to pay higher
wages. Magnus Blomström and Ari Kokko (both SSE) have published two
survey papers (one jointly with Steven Globerman (Simon Frazier
University)) on host-country spillovers from FDI.
FDI
and peripheral regions
For
the CSLA team, Barba Navaretti, Galeotti and Mattozzi examined the link
between imported technologies, through FDI and imports of machines, and
export performance. The analysis is set against the background of the
process of regional integration between the EU and its neighbouring
developing countries. The underlying question is whether trade integration
fosters or dampens learning and technological upgrading. The results show
that FDIs have a significant robust and positive impact on export
performance. The role of imported machines is more ambiguous. A
classification of the technological sophistication of imported machines
based on the minimum skills necessary to use them is developed. The
relationship between technological sophistication and export performance
is positive only when countries have a minimum skill base. This result
supports the hypothesis that skills are technology specific and that a
process of learning is necessary to achieve the positive virtuous circle
between imported technologies and performance.
The
same field of analysis has been followed by Barba Navaretti and Soloaga.
In particular, they examine the impact of imported technologies on
productivity for a sample of developing and transition countries in
Central and Eastern Europe and in the Southern Mediterranean. The results
show a constant and even increasing gap between the unit value of the
machines imported by the US and the machines imported by the sample of
developing countries. The empirical analysis also finds that productivity
in manufacturing depends positively on the type of machines imported in a
given industry. Consequently, although the choice of developing countries
to buy cheaper and less-sophisticated machines is optimal, given relative
factor prices and their endowments of technology, this choice has a cost
in terms of long-run productivity growth.
The
main focus of the Irish team has been on the impact of FDI on the Irish
economy, but Irish economists are increasingly seeking to locate their
experience within the context of the EU periphery as a whole, which is
taken to embrace Ireland, Greece, Spain and Portugal. As part of this
process, Salvador Barrios and Eric Strobl (both UCD) produced ‘FDI
Spillovers in Spain’. A further development of this research is the
paper by Barrios, Sophia Dimelis (IMOP), Helen Louri (IMOP) and Strobl
entitled ‘Foreign Direct Investment and Efficiency Spillovers in the EU
Periphery: A Comparative Study of Greece, Ireland and Spain’. This study
creates comparable data sets and estimates equivalent models for the three
peripheral EU economies and finds evidence of spillovers only for Ireland
and Spain. Positive spillovers seem to depend on whether firms have the
absorptive capacity to capture spillovers as well as on the exact
specification of foreign ownership.
In
the paper ‘Explaining Firms' Export Behaviour: The Role of R&D and
Spillovers’, Barrios (UCD), Görg (Nottingham) and Strobl (UCD) employ
Spanish data to show that export and R&D spillovers, whether from MNCs
or domestic firms, have different impacts on Spanish and on foreign firms,
with the latter generally benefiting from positive spillovers.
Following
Krugman, much recent work has been devoted to analysing similarities and
differences in industrial structures across countries. It is not clear as
yet, however, what the precise importance of such similarities and
differences might be, other than as indicators of economies’ asymmetric
vulnerabilities to sectoral shocks. In ‘FDI
and Structural Convergence in the EU Periphery’, Salvador Barrios, Frank
Barry and Eric Strobl (all UCD) focus
on the EU cohesion countries of Greece, Spain, Portugal and Ireland to
explore another dimension of the importance of structural similarities and
differences. The paper shows that convergence in industrial structure is
associated with convergence in terms of income per head, and assesses the
contribution of foreign industry in generating this structural convergence
amongst the traditionally poorer EU member states.
Frank
Barry also produced ‘Economic Policy, Income Convergence and Structural
Change in the EU Periphery’, which seeks to establish the types of
policies that can lead the periphery to converge on the more developed
core, and compares the extent to which the various EU periphery economies
have followed such policies over recent decades. Ireland’s FDI-based
strategy is argued to have been crucial to the rapidity of the country’s
recent convergence, and this is illustrated by demonstrating the role that
foreign industry played in driving the structural changes identified as
being associated with a move away from economic peripherality. The paper
argues that an FDI-driven development strategy such as Ireland’s
requires careful analysis in order to establish the real depth of
structural convergence, however.
Trade
integration leads to adjustment of either the inter- or intra-industry
variety. Factor endowments in the relatively poor economies of the EU
periphery would be expected to differ substantially from those at the
core, and so the trade-integration effects of the EU's Single Market
programme could be expected to lead to inter-industry adjustment in these
economies, in the form of an increase in specialisation on their part in
labour-intensive industries. In
‘Distorted Labour Markets and Revealed Comparative Advantage: A Note on
the Single Market and the EU Periphery’ Frank Barry and Aoife Hannan
show that indicators of revealed comparative advantage from the mid-1980s
predicted exactly this for Greece and Portugal, but predicted the opposite
for Ireland and Spain. In the Irish case at least this might be thought to
be the case because of the role of absolute advantage (in the form of very
low corporation tax rates) in attracting capital-intensive FDI. The
paradox remains, however, even when these sectors are excluded. The
results for Ireland and Spain are consistent with a model of Brecher’s,
however, in which labour markets are distorted. The paper argues that the
very high levels of unemployment that prevailed in Ireland and Spain at
the time of the Single Market are indicative of the presence of such
distortions.
The
IMOP team have also done work in this area. In ‘FDI in the EU Periphery:
A Multinomial Logit Analysis of Greek Firm Strategies’, Helen Louri,
Marina Papanastassiou and J Lantouris analyse the decision-making process
of Greek firms undertaken before they embark on outward FDI. They find
that borrowing capacity, labour intensity, sales growth rate, relative
firm size and acquired familiarity with foreign markets contribute to the
strategic decision of investing abroad.
Henrik
Braconier (IUI, Stockholm) and Karolina Ekholm (NHH/SNF, SSE and LSE) have
worked on Swedish FDI in Central and Eastern. Their work concentrates on
the potential effects of an expansion of firms in the CEE region on
employment in the home country, Sweden, and other European countries. The
main conclusion from their paper is that there is some evidence of a
relocation of employment within Swedish multinationals from the current
low-wage countries within the EU (Greece, Portugal, and Spain) to
countries in the CEE region.
Finally,
Guilia Faggio, a young researcher in the NHH/SNF team, has worked
extensively on FID, MNCs and CEECs. Among the questions she has addressed
are the determinants of the location decisions of MNCs in Central and
Eastern Europe, and the link between wages and FDI in transition
economies.
Determinants
of the location of FDI
The
location of FDI is often the outcome of a bidding game between countries
for the location of a subsidiary of a multinational firm. In the paper
‘On the Welfare Effects of the Competition for FDI’, Chiara Fumagalli
(UPF) analyses the location determinants of a multinational when two
counties bid for its services. Another
UPF team member, Enrico Pennings, has written three papers joint with Leo
Sleuwaegen (Catholic University of Leuven) on issues of delocalization.
The authors find that labour-intensive firms in a highly-industrialized
and open economy such as Belgium tend to relocate more to other countries
than their highly productive capital intensive counterparts.
Salvador
Barrios (UCD), Holger Görg (University of Ulster, University of
Nottingham and SSE), and Eric Strobl (UCD) have been working on policy
incentives and the location of multinationals. They look at Ireland, where
industrial policy-makers have offered explicit incentives for MNCs to
locate in the less-advantaged areas within the Republic in order to
eliminate regional disparities. In the resulting paper, the authors
estimate a nested multinomial logit of firm location to determine whether
these policy incentives have acted to influence the location of MNCs
within Ireland. Their empirical results find some support for this.
Helen
Louri (IMOP) has also carried out work in this area. In ‘Entry through
Acquisition: Determinants of Multinational Firm Choices’, she examines
the factors that determine the decision of MNCs to enter a foreign market
through acquisition. In addition to the traditional industry variables
attracting or discouraging entry through their effects on expected
returns, the relative size of MNC entry in Greece in 1987–96 is found to
be affected by a new group of variables, shaping rational profit
expectations and characterizing the target, the industry and the origin of
the buyer.
For
the CSLA team, Giorgio Barba Navaretti, Anna Falzoni and Alessandro
Turrini tested the firm-specific determinants of delocation to low-wage
countries on the part of Italian firms. The work is based on data
collected with a survey on a sample of enterprises in the textile and
clothing industries. These are two sectors where Italy has a strong
comparative advantage and which have re-deployed substantially. The
hypothesis, tested through a probit analysis, is that investments to cheap
labour countries are mainly cost-driven, and undertaken by firms that
focus on a low-quality, low-cost strategy. The evidence suggests that
investments to cheap labour countries are more likely to be of a vertical
type, being relatively more labour-intensive compared with the parent
company. The hypothesis seems to be confirmed empirically. Investments in
low-wage countries are more likely to generate abundant intra-firm trade
and to be undertaken by firms with low shares of skilled employment.
Using
a new panel dataset for Norwegian multinationals, the Norwegian team has
looked at characteristics of parent firms as well as subsidiaries, and
Norwegian MNCs’ motives for undertaking FDI, as reported, for example,
in the work carried out by Ingvild Selfors.
Building
on their theoretical work, Jan Haaland (NHH/SNF) and Ian Wooton
(University of Glasgow and CEPR) worked with Guilia Faggio (NHH/SNF) on
the role of flexible labour markets in attracting inward investment from
MNEs by adding empirical evidence to Haaland and Wooton’s earlier study
on the importance of labour flexibility for the attractiveness of a
location.
Karolina
Ekholm (SSE, NNH/SNF and LSE) has carried out an econometric analysis of
the determinants of intra-industry affiliate production, which is a
measure of the extent to which FDI is two-way directed within industries.
She finds support for the prevailing theory of FDI in-so-far as the extent
of intra-industry affiliate production is positively correlated with
similarity between countries in terms of their relative factor endowments
and market sizes.
In ‘Vertical FDI Revisited’, Henrik Braconier (IUI, Stockholm), Pehr-Johan Norbaek (IUI, Stockholm) and Dieter Urban (CSLA and LSE) explore how relative skilled-wage premia affect FDI. Contrary to previous studies based on factor endowment differences, they find strong support for vertical FDI, in the sense that more FDI is conducted in countries where unskilled labour is relatively cheap. In addition, the relative skill-premia also affect FDI activities that have previously been associated with horizontal FDI, i.e. local affiliate sales. Consequently, the potential effects of changes in the relative wage costs on international production reallocation within MNEs are large. In fact, if not for the 8% rise in the US skilled wage premium relative to the average host country between 1986-1994, annual US affiliate sales abroad in relation to US GDP would have been half a percentage point higher.
Return to Introduction
Research