Horizontal vs. Vertical Interdependence in Multinational Activity*
| Date | 01 December 2010 |
| Published date | 01 December 2010 |
| Author | Harald Badinger,Peter Egger |
| DOI | http://doi.org/10.1111/j.1468-0084.2010.00600.x |
744
©Blackwell Publishing Ltd and the Department of Economics, University of Oxford, 2010. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 72, 6 (2010) 0305-9049
doi: 10.1111/j.1468-0084.2010.00600.x
Horizontal vs. Vertical Interdependence in
Multinational ActivityÅ
Harald Badinger† and Peter Egger‡
†Vienna University of Economics and Business, Department of Economics, Augasse 2-6,
A-1090 Vienna (e-mail: harald.badinger@wu.ac.at)
‡ETH Zürich, KOF, WEH E6 Weinbergstrasse 35, 8092 Zürich, Switzerland
(e-mail: egger@kof.ethz.ch)
Abstract
This paper sheds light on interdependencies in multinational activity that are brought about
by (horizontal) trade in final goods and (vertical) trade in intermediate goods (within and
between host countries). Weuse a panel data set of US foreign affiliate sales to 16 developed
countries in seven industries over the period 1983–2000, distinguish between horizontal
and vertical interdependence in multinational enterprise activity and allow for both market
size (demand)-related as well as remainder linkage effects. Evidence suggests that vertical
interdependence is somewhat more important than horizontal interdependence and, hence,
vertical motives of multinational activity tend to dominate horizontal ones.
“. . . both export-platform and complex-vertical motivations imply that FDI decisions are
multilateral in nature . . . ” (Blonigen et al., 2007, p. 1304)
I. Introduction
A new strand of literature on the determinants of multinational enterprise (MNE) activity
in general and foreign direct investment (FDI) in particular highlights the interdependence
of MNEs’ investments and production across countries. Taking this interdependence into
account seems relevant for two reasons. First, empirical work on the determinants of MNE
activity mainly uses bilateral data and, in general equilibrium, bilateral flows depend not
only on bilateral but also on third-country determinants.1Second, evidence on the relative
importance of alternative channels of interdependence – e.g. trade in final goods vs.
trade in intermediates – is informative about the relevance of different motives of MNE
activity.
ÅWethank the editor in charge (Christopher Adam) and two anonymous referees for numerous helpful comments
on earlier versions of this paper.
JEL Classification numbers: C21, C23, F21, F23.
1The role of bilateral as well as third-country variables is well established in the literature on the determinants of
bilateral trade flows in so-called gravity models (see Eaton and Kortum, 2002; Anderson and van Wincoop, 2003).
However, while interdependence may be treated in a closed form in gravity models for goods flows, current models
of both MNE activity and trade only deliver reduced-form specifications. Accordingly, our contribution belongs to
the latter type of research.
Interdependence in multinational activity 745
Theoretical work establishes interdependence across markets explicitly through the
modelling of complex integration strategies of MNEs in more than two countries (see
Yeaple, 2003; Ekholm, Forslid and Markusen 2007; Grossman, Helpman, and Szeidl,
2007). There, integration strategies of multinationals are complex already by the number of
opportunities for locating production facilities and exporting goods bilaterally.Two-country
models of MNEs served to formalize stylized modes of MNE activity: horizontal and
vertical MNE organizations. Ahorizontal MNE produces the same good in either country
and does not engage in final goods trade (see Markusen, 1984; Markusen and Venables,
2000); a vertical MNE fully disentangles headquarters services from final goods produc-
tion, where the latter is located in the low-wage country with a comparative advantage in
goods production (see Helpman, 1984; Helpman and Krugman, 1985). With more than two
countries, complex MNEs arise, which are hybrid forms of horizontal and vertical MNEs.
They serve a subset of potential host countries through local foreign affiliates and others via
final goods trade from an export-platform subsidiary in a third host country (see Ekholm
et al., 2007) or even from a production site which is attached to the headquarters in the
parent country (see Egger, Egger and Ryan, 2007). The main channel of interdependence
in the aforementioned work is demand for final goods. On the one hand, foreign affi-
liate sales (FAS) to a potential host country may decline if its neighbours grow ceteris
paribus because this increases the home market and size of other firms’ subsidiaries
located there. On the other hand, FAS to that country may grow, if subsidiaries in that
market are used as a platform to serve consumers in a neighbouring country as well. This
form of interdependence across host markets roots in final-goods-demand-linkages across
markets.
Another strand of theoretical work emphasizes the role of vertically organized produc-
tion networks within MNEs (see Helpman, 1985; Markusen, 2002, ch. 7; Grossman et al.,
2007). Such networks establish an additional layer of interdependence through intra-firm
intermediate goods demand rather than inter-firm final goods demand.2The main mech-
anism at work there is that growing demand for an MNE’s final goods creates demand for
intermediate goods in all (relevant) subsidiaries within a firm.
An interesting feature of these two strands of theoretical work is that they provide
hypotheses about two channels of interdependence across host markets: one related to
trade in final goods and the other one associated with trade in intermediate goods.
Previous empirical work mainly considered final-goods-demand-related interdepen-
dence across host countries (see Baltagi, Egger and Pfaffermayr, 2007; Blonigen et al.,
2007) and identified a positive impact of third-country market size (‘market potential’).
Moreover, associated work found interdependences with respect to both observable and
unobservable variables captured by the disturbances in the empirical model (see Baltagi
et al., 2007; Blonigen et al., 2007). Overall, evidence has been interpreted as pointing to
the activity of complex MNEs in general and to export-platform activity in particular.
The goal of this paper is to explicitly disentangle vertical and horizontal modes of
interdependence in an empirical panel data model of the determinants of US FAS in 16
European countries. Sales of US MNEs’ foreign affiliates in these countries depend on
2In Helpman (1985), Markusen (2002, ch. 7) and Grossman et al. (2007), intermediate goods demand is strictly
within firms. However, a similar type of complementarities across host markets may arise with arms-length (i.e.
outside the firm) trade in intermediate goods as in Bergstrand and Egger (2008).
©Blackwell Publishing Ltd and the Department of Economics, University of Oxford 2010
To continue reading
Request your trial