Horizontal FDI
Horizontal multinationals are firms that produce the same good or services
in multiple plants in different countries, where each plant serves the local
market from the local production. Two factors are important for the appearance
of horizontal FDI: presence of positive trade costs and firm-level
scale economies. The main motivation for horizontal FDI is to avoid transportation
costs or to get access to a foreign market which can only be served
locally. The horizontal models predict that multinational activities can arise
between similar countries.
The intuition behind horizontal FDI is best described in form of an equation
with costs on the one side and benefits on the other side. Establishing a
foreign production instead of serving the market by exports means additional
costs of dealing with a new country. Moreover, there are production costs,
both fixed and variable, depending on factor prices and technology. The
plant-level economies of scales will increase the costs of establishing foreign
plants. On the other side of the equation, there are cost savings by switching
from exports to local production. The most obvious are transport costs and
tariffs. Additional benefits arise from the proximity to the market, as shorter
delivery and quicker response to the market becomes easier. Thus, if benefits
outweigh the costs a multinational enterprise will conduct a horizontal FDI.
The models of horizontal FDI predict, that given the existence of trade
costs and economies of scale at plant and firm level, investment flows can
arise between similar countries (see Table 2.1)
In order to explain the intuition
behind the models we look at the counter-example with two countrieswhich are different in either size or in relative factor endowments. In both
situations we assume moderate transportation costs. In the first case, the
appearance of horizontal multinationals is unlikely, because they will have a
disadvantage relative to the national firm with headquarter and production
plant in the larger country. The multinational has to bear fixed costs for
the plant in the smaller market, while the national firm in the larger country
faces trade costs for the small amount of exports to the smaller country. In
the second case the countries are similar in size but different in factor endowments.
The horizontal multinational has a disadvantage again, since it places
the production in both countries, also in the more expensive, factor-scarce
country. The national firm located in the country which is factor-abundant
(for example with labour force) conducts the complete production in the
low cost country. The presence of transportation costs is thereby important,
since otherwise the foreign markets will be served by exports and the firm
uses only the scale effects by setting the complete production in one plant.
Theoretical models of horizontal FDI are based on the trade-off between
additional fixed costs from setting up a new plant and the saving of variable
costs from avoiding tariffs and transportation. One of the earliest models
on horizontal FDI is Markusen (1984) with firm-level scale economies as a
driving force. A two-plant firm has fixed costs that are less than double the
ones of a single plant firm, thus creating a motivation for multi-plant production.
Extensions and refinements of this model can be found in Horstman
and Markusen (1987, 1992). Markusen (1995) provides a discussion on horizontal
MNE as an alternative to trade and local firms and discusses the
internalisation problem. A survey of literature and an overview of empirical
finding can be found in Markusen and Maskus (2001).
In a more general model Brainard (1993) discusses the role of scale effects
at the firm and plant level in relation to transportation costs. The intuition is
that horizontal FDI appears as an alternative to exports, if the trade costs are
larger than the fixed costs from establishing a new plant, which is also known
under the term “proximity-concentration approach”. The driving force here
is the trade-off between the advantages of being near to the market to avoid
transportation costs (proximity) and scale effects in case of production in
one plant (concentration). Scale effects occur because of fixed costs when
building a new plant. The model predicts two situations when horizontal
FDI will dominate over exports or crowd them out completely. The first is
when the transportation costs are large in comparison to the plant fixed costs,
while the second occurs when firm-level scale effects are larger than plantlevel
scale effects. This means that the incentive for horizontal multinationals
increases the greater are transport costs relative to fixed plant costs and the
greater are increasing returns at the firm level relative to the plant level.
Further developments of the horizontal model were conducted by Markusen
and Venables (1998, 2000). Markusen and Venables (1998) extended the
aforementioned models to a full multi-country framework, allowing for the
mix of multinational and local firms in each country. In the former, multinationals
dominate in countries that are similar in size, factor and technology
endowments. In the latter, the authors show that dissimilarity in relative
factor endowments reduce the horizontal activity of MNE.
Table 2.1 summarizes the conditions necessary for the appearance of horizontal
multinationals: countries similar in size and factor endowments, presence
of transportation costs and economies of scale at the firm level.
The findings from the models of horizontal FDI can explain a variety of
features of FDI flows. First, horizontal FDI reduces trade flows, since the
market is served through local production instead of exports. Second, horizontal
FDI takes place if the costs of importing are high relative to costs
of investing. Third, horizontal FDI is more likely to occur in large foreign
markets, which allows to spread fixed costs for the new plant over a large
volume of production. Finally, the value of local production may exceed the
simple calculation of net costs from the described trade-off, when establishing
a local production plant may have a strategic value. Given an oligopolistic
market, sales of each firm depend on the marginal costs of all other competitors.
By conducting horizontal FDI the firm reduces its marginal costs, what
may induce the other firms to reduce their sales. Setting up a new plant is
also a commitment to supply the local market, and this commitment may
change the behaviour of competitors.