Multinationals and Outsourcing
When is a corporation multinational? In U.S. statistics, a U.S. company is considered foreign-
controlled, and therefore a subsidiary of a foreign-based multinational, if 10 percent
or more of its stock is held by a foreign company; the idea is that 10 percent is enough to
convey effective control. Similarly, a U.S.-based company is considered multinational if it
owns more than 10 percent of a foreign firm. The controlling (owning) firm is called the
multinational parent, while the “controlled” firms are called the multinational affiliates.
When a U.S. firm buys more than 10 percent of a foreign firm, or when a U.S. firm
builds a new production facility abroad, that investment is considered a U.S. outflow of
foreign direct investment (FDI). The latter is called greenfield FDI, while the former is
called brownfield FDI (or cross-border mergers and acquisitions). Conversely, investments
by foreign firms in production facilities in the United States are considered U.S. FDI
inflows. We describe the worldwide patterns of FDI flows in the Case Study that follows.
For now, we focus on the decision of a firm to become a multinational parent. Why would
a firm choose to operate an affiliate in a foreign location?
18“Wielding a Heavy Weapon Against China,” Business Week, June 21, 2004.
Case Study
Patterns of Foreign Direct Investment Flows Around the World
Figure 8-9 shows how the magnitude of worldwide FDI flows has evolved over the last
30 years. We first examine patterns for the world, where FDI flows must be balanced:
Hence world inflows are equal to world outflows. We see that there was a massive increase
in multinational activity in the mid- to late 1990s, when worldwide FDI flows
more than quintupled, and then again in the early 2000s. We also see that the growth
rate of FDI is very uneven, with huge peaks and troughs. Those peaks and troughs
correlate with the gyrations of stock markets worldwide (strongly dominated by fluctuations
in the U.S. stock market). The financial collapse in 2000 (the bursting of the dotcom
bubble) and the most recent financial crisis in 2007–2009 also induced huge
crashes in worldwide FDI flows. Most of those FDI flows related to cross-border mergers
and acquisitions, whereas greenfield FDI remained relatively stable.
CHAPTER 8 Firms in the Global Economy 181
0
500,000
1,000,000
2,500,000
1,500,000
2,000,000
FDI inflows
(billions of dollars)
1980 1985 1990 1995 2000 2005 2010
Transition economies
Developed economies
Developing economies
Figure 8-9
Inflows of Foreign Direct Investment, 1980–2009 (billions of dollars)
Worldwide flows of FDI have significantly increased since the mid-1990s, though the rates of increase
have been very uneven. Historically, most of the inflows of FDI have gone to developed countries.
However, the proportion of FDI inflows going to developing and transition economies has steadily
increased over time and accounted for half of worldwide FDI flows in 2009.
Source: UNCTAD, World Investment Report, 2010.
Looking at the distribution of FDI inflows across groups of countries, we see that historically,
developed countries have been the biggest recipients of inward FDI. However,
we also see that those inflows are much more volatile (this is where the FDI related to
mergers and acquisitions is concentrated) than the FDI going to developing and transition
economies (economies in Central/Eastern Europe that used to be part of the Soviet Union
or Yugoslavia). Finally, we can see that there has been a steady expansion in the share of
FDI that flows to developing and transition countries. This accounted for half of worldwide
FDI flows in 2009, after the most recent contraction in the flows to developed economies.
Figure 8-10 shows the list of the top 25 countries whose firms engage in FDI outflows.
Because those flows are very volatile, especially with the recent crisis, they have
been averaged over the past three years. We see that FDI outflows are still dominated by
the developed economies; but we also see that big developing countries, most notably
China (including Hong Kong), are playing an increasingly important role. In fact, one of
the fastest-growing FDI segments is flows from developing countries into other developing
countries. Multinationals in both China and India play a prominent role in this relatively
new type of FDI. We also see that international tax policies can shape the location
of FDI. For example, the British Virgin Islands would not figure in that top-25 list were
it not for its status as an international tax haven. Firms from that location that engage in
FDI are mainly offshore companies: They are incorporated in the British Virgin Islands,
but their productive activities are located elsewhere in the world.
FDI flows are not the only way to measure the presence of multinationals in the
world economy. Other measures are based on economic activities such as sales, value
182 PART ONE International Trade Theory
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
FDI outflows
(billions of dollars)
United States
United Kingdom
France
Germany
Japan
Spain
Belgium
Canada
Italy
China, Hong Kong
Russian Federation
Switzerland
China
Hungary
Sweden
Luxembourg
British Virgin Islands
Norway
Austria
Australia
Netherlands
Ireland
India
Denmark
Korea, Republic of
Figure 8-10
Outward Foreign Direct Investment for Top 25 Countries, Yearly Average for 2007–2009 (billions of dollars)
Developed countries dominate the list of the top countries whose firms engage in outward FDI. More recently,
firms from some big developing countries such as China and India have performed significantly more FDI.
Source: UNCTAD, World Investment Report, 2010.
added (sales minus purchased intermediate goods), and employment. Sales of FDI affiliates
are often used as the benchmark of multinational activity. This provides the relevant
benchmark when comparing the activities of multinationals to export volumes.
However, the sales of multinationals are also often compared to country GDPs showing,
for example, that the big multinationals have higher sales volumes than the GDPs
of many countries in the world. For the world as a whole in 2000, the total sales of the
largest multinationals (top 200) amounted to more than 27 percent of world GDP.
However striking, this comparison is misleading and overstates the influence of
multinationals, because country GDP is measured in terms of value added: Intermediate
goods used in final production are not double-counted in this GDP measure. On the
other hand, the intermediate goods that one multinational sells to another are doublecounted
in the multinationals’ sales totals (once in the sales of the producer of the intermediate
goods, and another time as part of the final value of the goods sold by the user
of the intermediate goods). As a result, the appropriate comparison between multinationals
and GDPs should be based on value added. By this metric, the value added produced
by the biggest multinationals accounted for 4.3 percent of world GDP in 2000.
This is still a big percentage, but not as eye-catching as the 27 percent measure.
CHAPTER 8 Firms in the Global Economy 183
The answer depends, in part, on the production activities that the affiliate carries out.
These activities fall into two main categories: (1) The affiliate replicates the production
process (that the parent firm undertakes in its domestic facilities) elsewhere in the world;
and (2) the production chain is broken up, and parts of the production processes are transferred
to the affiliate location. Investing in affiliates that do the first type of activities is categorized
as horizontal FDI. Investing in affiliates that do the second type of activities is
categorized as vertical FDI.19
Vertical FDI is mainly driven by production cost differences between countries (for
those parts of the production process that can be performed in another location). What
drives those cost differences between countries? This is just the outcome of the theory of
comparative advantage that we developed in Chapters 3 through 7. For example, Intel (the
world’s largest computer chip manufacturer) has broken up the production of chips into
wafer fabrication, assembly, and testing. Wafer fabrication and the associated research and
development are very skill-intensive, so Intel still performs most of those activities in the
United States, as well as in Ireland and Israel (where skilled labor is still relatively abundant).
On the other hand, chip assembly and testing are labor-intensive, and Intel has
moved those production processes to countries where labor is relatively abundant, such as
Malaysia, the Philippines, and, more recently, Costa Rica and China. This type of vertical
FDI is one of the fastest-growing types of FDI, and is behind the large increase in FDI inflows
to developing countries (see Figure 8-9).
In contrast to vertical FDI, horizontal FDI is dominated by flows between developed
countries; that is, both the multinational parent and the affiliates are located in developed
countries. The main reason for this type of FDI is to locate production near a firm’s large
customer bases. Hence, trade and transport costs play a much more important role than production
cost differences for these FDI decisions. Consider the example of Toyota, which is
the world’s largest motor vehicle producer (at least, at the time of writing). At the start of
the 1980s, Toyota produced almost all of its cars and trucks in Japan and exported them
throughout the world, but mostly to North America and Europe. High trade costs to those
markets (in large part due to trade restrictions; see Chapter 9) and rising demand levels
there induced Toyota to slowly expand its production overseas. By 2009, Toyota produced
over half of its vehicles in assembly plants abroad. Toyota has replicated the production
process for its most popular car model, the Corolla, in assembly plants in Japan, Canada,
the United States, the United Kingdom, and Turkey: This is horizontal FDI in action.
The Firm’s Decision Regarding Foreign Direct Investment
We now examine in more detail the firm’s decision regarding horizontal FDI. We mentioned
that one main driver was high trade costs associated with exporting, which leads
to an incentive to locate production near customers. On the other hand, there are also
increasing returns to scale in production. As a result, it is not cost effective to replicate
the production process too many times and operate facilities that produce too little output
to take advantage of those increasing returns. This is called the proximity-concentration
trade-off for FDI. Empirical evidence on the extent of FDI across sectors strongly confirms
the relevance of this trade-off: FDI activity is concentrated in sectors where trade
costs are high (such as the automobile industry); however, when increasing returns to
scale are important and average plant sizes are large, one observes higher export volumes
relative to FDI.
19In reality, the distinctions between horizontal and vertical FDI can be blurred. Some large multinational parents
operate large networks of affiliates that replicate parts of the production process, but are also vertically connected
to other affiliates in the parent’s network. This is referred to as “complex” FDI.
184 PART ONE International Trade Theory
Empirical evidence also shows that there is an even stronger sorting pattern for FDI at
the firm level within industries: Multinationals tend to be substantially larger and more
productive than nonmultinationals in the same country. Even when one compares multinationals
to the subset of exporting firms in a country, one still finds a large size and productivity
differential in favor of the multinationals. We return to our monopolistic competition
model of trade to analyze how firms respond differently to the proximity-concentration
trade-off involved with the FDI decision.
The Horizontal FDI Decision How does the proximity trade-off fit into our model of
firms’ export decisions captured in Figure 8-8? There, if a firm wants to reach customers
in Foreign, it has only one possibility: export, and incur the trade cost t per unit exported.
Let’s now introduce the choice of becoming a multinational via horizontal FDI: A firm
could avoid the trade cost t by building a production facility in Foreign. Of course,
building this production facility is costly, and implies incurring the fixed cost F again for
the foreign affiliate. (Note, however, that this additional fixed cost need not equal the fixed
cost of building the firm’s original production facility in Home; characteristics that are
specific to the individual country will affect this cost.) For simplicity, continue to assume
that Home and Foreign are similar countries so that this firm could build a unit of a good at
the same marginal cost in this foreign facility. (Recall that horizontal FDI mostly involves
developed countries with similar factor prices.)
The firm’s export versus FDI choice will then involve a trade-off between the per-unit
export cost t and the fixed cost F of setting up an additional production facility. Any such
trade-off between a per-unit and a fixed cost boils down to scale. If the firm sells Q units in
the foreign market, then it incurs a total trade-related cost to export; this is weighed
against the alternative of the fixed cost F. If , then exporting is more expensive,
and FDI is the profit-maximizing choice.
This leads to a scale cutoff for FDI. This cutoff summarizes the proximity-concentration
trade-off: Higher trade costs on one hand, and lower fixed production costs on the other
hand, both lower the FDI cutoff. The firm’s scale, however, depends on its performance
measure. A firm with low enough cost will want to sell more than Q units to foreign customers.
The most cost-effective way to do this is to build an affiliate in Foreign and become
a multinational. Some firms with intermediate cost levels will still want to serve customers
in Foreign, but their intended sales Q are low enough that exports, rather than FDI, will be
the most cost-effective way to reach those customers.
The Vertical FDI Decision A firm’s decision to break up its production chain and move
parts of that chain to a foreign affiliate will also involve a trade-off between per-unit and
fixed costs—so the scale of the firm’s activity will again be a crucial element determining
this outcome. When it comes to vertical FDI, the key cost saving is not related to the
shipment of goods across borders; rather, it involves production cost differences for the
parts of the production chain that are being moved. As we previously discussed, those cost
differences stem mostly from comparative advantage forces.
We will not discuss those cost differences further here, but rather ask why—given those
cost differences—all firms do not choose to operate affiliates in low-wage countries to perform
the activities that are most labor-intensive and can be performed in a different location.
The reason is that, as with the case of horizontal FDI, vertical FDI requires a substantial
fixed cost investment in a foreign affiliate in a country with the appropriate characteristics.20
ci
Q 7 F/t
Q * t
20Clearly, factor prices such as wages are a crucial component, but other country characteristics, such as its
transportation/public infrastructure, the quality of its legal institutions, and its tax/regulation policies toward
multinationals, can be critical as well.
CHAPTER 8 Firms in the Global Economy 185
Again, as with the case of horizontal FDI, there will be a scale cutoff for vertical FDI that
depends on the production cost differentials on one hand, and the fixed cost of operating a
foreign affiliate on the other hand. Only those firms operating at a scale above that cutoff will
choose to perform vertical FDI.
Outsourcing
Our discussion of multinationals up to this point has neglected an important motive. We
discussed the location motive for production facilities that leads to multinational formation.
However, we did not discuss why the parent firm chooses to own the affiliate in that
location and operate as a single multinational firm. This is known as the internalization
motive.
As a substitute for horizontal FDI, a parent could license an independent firm to produce
and sell its products in a foreign location; as a substitute for vertical FDI, a parent
could contract with an independent firm to perform specific parts of the production
process in the foreign location with the best cost advantage. This substitute for vertical
FDI is known as foreign outsourcing (sometimes just referred to as outsourcing, where
the foreign location is implied).
Offshoring represents the relocation of parts of the production chain abroad and
groups together both foreign outsourcing and vertical FDI. Offshoring has increased dramatically
in the last decade and is one of the major drivers of the increased worldwide
trade in services (such as business and telecommunications services); in manufacturing,
trade in intermediate goods accounted for 40 percent of worldwide trade in 2008. When
the intermediate goods are produced within a multinational’s affiliate network, the shipments
of those intermediate goods are classified as intra-firm trade. Intra-firm trade represents
roughly one-third of worldwide trade and over 40 percent of U.S. trade.
What are the key elements that determine this internalization choice? Control over a
firm’s proprietary technology offers one clear advantage for internalization. Licensing another
firm to perform the entire production process in another location (as a substitute for
horizontal FDI) often involves a substantial risk of losing some proprietary technology. On
the other hand, there are no clear reasons why an independent firm should be able to replicate
that production process at a lower cost than the parent firm. This gives internalization
a strong advantage, so horizontal FDI is widely favored over the alternative of technology
licensing to replicate the production process.
The trade-off between outsourcing and vertical FDI is much less clear-cut. There are
many reasons why an independent firm could produce some parts of the production
process at lower cost than the parent firm (in the same location). First and foremost, an
independent firm can specialize in exactly that narrow part of the production process. As a
result, it can also benefit from economies of scale if it performs those processes for many
different parent firms.21 Other reasons stress the advantages of local ownership in the
alignment and monitoring of managerial incentives at the production facility.
But internalization also provides its own benefits when it comes to vertical integration
between a firm and its supplier of a critical input to production: This avoids (or at least
lessens) the potential for a costly renegotiation conflict after an initial agreement has been
reached. Such conflicts can arise regarding many specific attributes of the input that cannot
be specified in (or enforced by) a legal contract written at the time of the initial agreement.
This can lead to a holdup of production by either party. For example, the buying firm can
21Companies that provide outsourced goods and services have expanded their list of clients to such an extent that
they have now become large multinationals themselves. They specialize in providing a narrow set of services (or
parts of the production process), but replicate this many times over for client companies across the globe.
186 PART ONE International Trade Theory
claim that the quality of the part is not exactly as specified and demand a lower price. The
supplying firm can claim that some changes demanded by the buyer led to increased costs
and demand a higher price at delivery time.
Much progress has been made in recent research formalizing those trade-offs. This research
explains how this important internalization choice is made, by describing when a
firm chooses to integrate with its suppliers via vertical FDI and when it chooses an independent
contractual relationship with those suppliers abroad. Developing those theories is
beyond the scope of this textbook; ultimately, many of those theories boil down to different
trade-offs between production cost savings and the fixed cost of moving parts of the
production process abroad.
Describing which types of firms pick one offshoring option versus the other is sensitive
to the details of the modeling assumptions. Nonetheless, one robust prediction emerges
from those models when one compares either offshoring option to that of no offshoring
(not breaking up the production chain and moving parts of it abroad). Relative to no offshoring,
both vertical FDI and foreign outsourcing involve lower production costs combined
with a higher fixed cost. As we saw, this implies a scale cutoff for a firm to choose
either offshoring option. Thus, only the larger firms will choose either offshoring option
and import some of their intermediate inputs.
This sorting scheme for firms to import intermediate goods is similar to the one we
described for the firm’s export choice: Only a subset of relatively more productive (lowercost)
firms will choose to offshore (import intermediate goods) and export (reach foreign
customers)—because those are the firms that operate at sufficiently large scale to favor the
trade-off involving higher fixed costs and lower per-unit costs (production- or traderelated).
Empirically, are the firms that offshore and import intermediate goods the same set of
firms that also export? The answer is a resounding yes. For the United States in 2000,
92 percent of firms (weighed by employment) that imported intermediate goods also
exported. Those importers thus also shared the same characteristics as U.S. exporters:
They were substantially larger and more productive than the U.S. firms that did not engage
in international trade.
Consequences of Multinationals and Foreign Outsourcing
Earlier in this chapter, we mentioned that internal economies of scale, product differentiation,
and performance differences across firms combined to deliver some new channels for
the gains from trade: increased product variety, and higher industry performance as firms
move down their average cost curve and production is concentrated in the larger, more
productive firms. What are the consequences for welfare of the expansion in multinational
production and outsourcing?
We just saw how multinationals and firms that outsource take advantage of cost differentials
that favor moving production (or parts thereof) to particular locations. In essence,
this is very similar to the relocation of production that occurred across sectors when opening
to trade. As we saw in Chapters 3 through 6, the location of production then shifts to
take advantage of cost differences generated by comparative advantage.
We can therefore predict similar welfare consequences for the case of multinationals
and outsourcing: Relocating production to take advantage of cost differences leads to
overall gains from trade, but it is also likely to induce income distribution effects that leave
some people worse off. We discussed one potential long-run consequence of outsourcing
for income inequality in developed countries in Chapter 5.
Yet some of the most visible effects of multinationals and outsourcing occur in the
short run, as some firms expand employment while others reduce employment in response
CHAPTER 8 Firms in the Global Economy 187
to increased globalization. We mentioned in Chapter 4 that those employment changes due
to overseas plant relocations (along with plant closures due to import competition) account
for only a small fraction (2.5 percent) of all involuntary worker displacements in the
United States. Nevertheless, when such plant relocations do occur, they inevitably generate
some substantial costs for those affected workers. As we argued in Chapter 4, the best
policy response to this serious concern is still to provide an adequate safety net to unemployed
workers without discriminating based on the economic force that induced their
involuntary unemployment. Policies that impede firms’ abilities to relocate production and
take advantage of these cost differences may prevent these short-run costs for some, but
they also forestall the accumulation of long-run economy-wide gains.
When is a corporation multinational? In U.S. statistics, a U.S. company is considered foreign-
controlled, and therefore a subsidiary of a foreign-based multinational, if 10 percent
or more of its stock is held by a foreign company; the idea is that 10 percent is enough to
convey effective control. Similarly, a U.S.-based company is considered multinational if it
owns more than 10 percent of a foreign firm. The controlling (owning) firm is called the
multinational parent, while the “controlled” firms are called the multinational affiliates.
When a U.S. firm buys more than 10 percent of a foreign firm, or when a U.S. firm
builds a new production facility abroad, that investment is considered a U.S. outflow of
foreign direct investment (FDI). The latter is called greenfield FDI, while the former is
called brownfield FDI (or cross-border mergers and acquisitions). Conversely, investments
by foreign firms in production facilities in the United States are considered U.S. FDI
inflows. We describe the worldwide patterns of FDI flows in the Case Study that follows.
For now, we focus on the decision of a firm to become a multinational parent. Why would
a firm choose to operate an affiliate in a foreign location?
18“Wielding a Heavy Weapon Against China,” Business Week, June 21, 2004.
Case Study
Patterns of Foreign Direct Investment Flows Around the World
Figure 8-9 shows how the magnitude of worldwide FDI flows has evolved over the last
30 years. We first examine patterns for the world, where FDI flows must be balanced:
Hence world inflows are equal to world outflows. We see that there was a massive increase
in multinational activity in the mid- to late 1990s, when worldwide FDI flows
more than quintupled, and then again in the early 2000s. We also see that the growth
rate of FDI is very uneven, with huge peaks and troughs. Those peaks and troughs
correlate with the gyrations of stock markets worldwide (strongly dominated by fluctuations
in the U.S. stock market). The financial collapse in 2000 (the bursting of the dotcom
bubble) and the most recent financial crisis in 2007–2009 also induced huge
crashes in worldwide FDI flows. Most of those FDI flows related to cross-border mergers
and acquisitions, whereas greenfield FDI remained relatively stable.
CHAPTER 8 Firms in the Global Economy 181
0
500,000
1,000,000
2,500,000
1,500,000
2,000,000
FDI inflows
(billions of dollars)
1980 1985 1990 1995 2000 2005 2010
Transition economies
Developed economies
Developing economies
Figure 8-9
Inflows of Foreign Direct Investment, 1980–2009 (billions of dollars)
Worldwide flows of FDI have significantly increased since the mid-1990s, though the rates of increase
have been very uneven. Historically, most of the inflows of FDI have gone to developed countries.
However, the proportion of FDI inflows going to developing and transition economies has steadily
increased over time and accounted for half of worldwide FDI flows in 2009.
Source: UNCTAD, World Investment Report, 2010.
Looking at the distribution of FDI inflows across groups of countries, we see that historically,
developed countries have been the biggest recipients of inward FDI. However,
we also see that those inflows are much more volatile (this is where the FDI related to
mergers and acquisitions is concentrated) than the FDI going to developing and transition
economies (economies in Central/Eastern Europe that used to be part of the Soviet Union
or Yugoslavia). Finally, we can see that there has been a steady expansion in the share of
FDI that flows to developing and transition countries. This accounted for half of worldwide
FDI flows in 2009, after the most recent contraction in the flows to developed economies.
Figure 8-10 shows the list of the top 25 countries whose firms engage in FDI outflows.
Because those flows are very volatile, especially with the recent crisis, they have
been averaged over the past three years. We see that FDI outflows are still dominated by
the developed economies; but we also see that big developing countries, most notably
China (including Hong Kong), are playing an increasingly important role. In fact, one of
the fastest-growing FDI segments is flows from developing countries into other developing
countries. Multinationals in both China and India play a prominent role in this relatively
new type of FDI. We also see that international tax policies can shape the location
of FDI. For example, the British Virgin Islands would not figure in that top-25 list were
it not for its status as an international tax haven. Firms from that location that engage in
FDI are mainly offshore companies: They are incorporated in the British Virgin Islands,
but their productive activities are located elsewhere in the world.
FDI flows are not the only way to measure the presence of multinationals in the
world economy. Other measures are based on economic activities such as sales, value
182 PART ONE International Trade Theory
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
FDI outflows
(billions of dollars)
United States
United Kingdom
France
Germany
Japan
Spain
Belgium
Canada
Italy
China, Hong Kong
Russian Federation
Switzerland
China
Hungary
Sweden
Luxembourg
British Virgin Islands
Norway
Austria
Australia
Netherlands
Ireland
India
Denmark
Korea, Republic of
Figure 8-10
Outward Foreign Direct Investment for Top 25 Countries, Yearly Average for 2007–2009 (billions of dollars)
Developed countries dominate the list of the top countries whose firms engage in outward FDI. More recently,
firms from some big developing countries such as China and India have performed significantly more FDI.
Source: UNCTAD, World Investment Report, 2010.
added (sales minus purchased intermediate goods), and employment. Sales of FDI affiliates
are often used as the benchmark of multinational activity. This provides the relevant
benchmark when comparing the activities of multinationals to export volumes.
However, the sales of multinationals are also often compared to country GDPs showing,
for example, that the big multinationals have higher sales volumes than the GDPs
of many countries in the world. For the world as a whole in 2000, the total sales of the
largest multinationals (top 200) amounted to more than 27 percent of world GDP.
However striking, this comparison is misleading and overstates the influence of
multinationals, because country GDP is measured in terms of value added: Intermediate
goods used in final production are not double-counted in this GDP measure. On the
other hand, the intermediate goods that one multinational sells to another are doublecounted
in the multinationals’ sales totals (once in the sales of the producer of the intermediate
goods, and another time as part of the final value of the goods sold by the user
of the intermediate goods). As a result, the appropriate comparison between multinationals
and GDPs should be based on value added. By this metric, the value added produced
by the biggest multinationals accounted for 4.3 percent of world GDP in 2000.
This is still a big percentage, but not as eye-catching as the 27 percent measure.
CHAPTER 8 Firms in the Global Economy 183
The answer depends, in part, on the production activities that the affiliate carries out.
These activities fall into two main categories: (1) The affiliate replicates the production
process (that the parent firm undertakes in its domestic facilities) elsewhere in the world;
and (2) the production chain is broken up, and parts of the production processes are transferred
to the affiliate location. Investing in affiliates that do the first type of activities is categorized
as horizontal FDI. Investing in affiliates that do the second type of activities is
categorized as vertical FDI.19
Vertical FDI is mainly driven by production cost differences between countries (for
those parts of the production process that can be performed in another location). What
drives those cost differences between countries? This is just the outcome of the theory of
comparative advantage that we developed in Chapters 3 through 7. For example, Intel (the
world’s largest computer chip manufacturer) has broken up the production of chips into
wafer fabrication, assembly, and testing. Wafer fabrication and the associated research and
development are very skill-intensive, so Intel still performs most of those activities in the
United States, as well as in Ireland and Israel (where skilled labor is still relatively abundant).
On the other hand, chip assembly and testing are labor-intensive, and Intel has
moved those production processes to countries where labor is relatively abundant, such as
Malaysia, the Philippines, and, more recently, Costa Rica and China. This type of vertical
FDI is one of the fastest-growing types of FDI, and is behind the large increase in FDI inflows
to developing countries (see Figure 8-9).
In contrast to vertical FDI, horizontal FDI is dominated by flows between developed
countries; that is, both the multinational parent and the affiliates are located in developed
countries. The main reason for this type of FDI is to locate production near a firm’s large
customer bases. Hence, trade and transport costs play a much more important role than production
cost differences for these FDI decisions. Consider the example of Toyota, which is
the world’s largest motor vehicle producer (at least, at the time of writing). At the start of
the 1980s, Toyota produced almost all of its cars and trucks in Japan and exported them
throughout the world, but mostly to North America and Europe. High trade costs to those
markets (in large part due to trade restrictions; see Chapter 9) and rising demand levels
there induced Toyota to slowly expand its production overseas. By 2009, Toyota produced
over half of its vehicles in assembly plants abroad. Toyota has replicated the production
process for its most popular car model, the Corolla, in assembly plants in Japan, Canada,
the United States, the United Kingdom, and Turkey: This is horizontal FDI in action.
The Firm’s Decision Regarding Foreign Direct Investment
We now examine in more detail the firm’s decision regarding horizontal FDI. We mentioned
that one main driver was high trade costs associated with exporting, which leads
to an incentive to locate production near customers. On the other hand, there are also
increasing returns to scale in production. As a result, it is not cost effective to replicate
the production process too many times and operate facilities that produce too little output
to take advantage of those increasing returns. This is called the proximity-concentration
trade-off for FDI. Empirical evidence on the extent of FDI across sectors strongly confirms
the relevance of this trade-off: FDI activity is concentrated in sectors where trade
costs are high (such as the automobile industry); however, when increasing returns to
scale are important and average plant sizes are large, one observes higher export volumes
relative to FDI.
19In reality, the distinctions between horizontal and vertical FDI can be blurred. Some large multinational parents
operate large networks of affiliates that replicate parts of the production process, but are also vertically connected
to other affiliates in the parent’s network. This is referred to as “complex” FDI.
184 PART ONE International Trade Theory
Empirical evidence also shows that there is an even stronger sorting pattern for FDI at
the firm level within industries: Multinationals tend to be substantially larger and more
productive than nonmultinationals in the same country. Even when one compares multinationals
to the subset of exporting firms in a country, one still finds a large size and productivity
differential in favor of the multinationals. We return to our monopolistic competition
model of trade to analyze how firms respond differently to the proximity-concentration
trade-off involved with the FDI decision.
The Horizontal FDI Decision How does the proximity trade-off fit into our model of
firms’ export decisions captured in Figure 8-8? There, if a firm wants to reach customers
in Foreign, it has only one possibility: export, and incur the trade cost t per unit exported.
Let’s now introduce the choice of becoming a multinational via horizontal FDI: A firm
could avoid the trade cost t by building a production facility in Foreign. Of course,
building this production facility is costly, and implies incurring the fixed cost F again for
the foreign affiliate. (Note, however, that this additional fixed cost need not equal the fixed
cost of building the firm’s original production facility in Home; characteristics that are
specific to the individual country will affect this cost.) For simplicity, continue to assume
that Home and Foreign are similar countries so that this firm could build a unit of a good at
the same marginal cost in this foreign facility. (Recall that horizontal FDI mostly involves
developed countries with similar factor prices.)
The firm’s export versus FDI choice will then involve a trade-off between the per-unit
export cost t and the fixed cost F of setting up an additional production facility. Any such
trade-off between a per-unit and a fixed cost boils down to scale. If the firm sells Q units in
the foreign market, then it incurs a total trade-related cost to export; this is weighed
against the alternative of the fixed cost F. If , then exporting is more expensive,
and FDI is the profit-maximizing choice.
This leads to a scale cutoff for FDI. This cutoff summarizes the proximity-concentration
trade-off: Higher trade costs on one hand, and lower fixed production costs on the other
hand, both lower the FDI cutoff. The firm’s scale, however, depends on its performance
measure. A firm with low enough cost will want to sell more than Q units to foreign customers.
The most cost-effective way to do this is to build an affiliate in Foreign and become
a multinational. Some firms with intermediate cost levels will still want to serve customers
in Foreign, but their intended sales Q are low enough that exports, rather than FDI, will be
the most cost-effective way to reach those customers.
The Vertical FDI Decision A firm’s decision to break up its production chain and move
parts of that chain to a foreign affiliate will also involve a trade-off between per-unit and
fixed costs—so the scale of the firm’s activity will again be a crucial element determining
this outcome. When it comes to vertical FDI, the key cost saving is not related to the
shipment of goods across borders; rather, it involves production cost differences for the
parts of the production chain that are being moved. As we previously discussed, those cost
differences stem mostly from comparative advantage forces.
We will not discuss those cost differences further here, but rather ask why—given those
cost differences—all firms do not choose to operate affiliates in low-wage countries to perform
the activities that are most labor-intensive and can be performed in a different location.
The reason is that, as with the case of horizontal FDI, vertical FDI requires a substantial
fixed cost investment in a foreign affiliate in a country with the appropriate characteristics.20
ci
Q 7 F/t
Q * t
20Clearly, factor prices such as wages are a crucial component, but other country characteristics, such as its
transportation/public infrastructure, the quality of its legal institutions, and its tax/regulation policies toward
multinationals, can be critical as well.
CHAPTER 8 Firms in the Global Economy 185
Again, as with the case of horizontal FDI, there will be a scale cutoff for vertical FDI that
depends on the production cost differentials on one hand, and the fixed cost of operating a
foreign affiliate on the other hand. Only those firms operating at a scale above that cutoff will
choose to perform vertical FDI.
Outsourcing
Our discussion of multinationals up to this point has neglected an important motive. We
discussed the location motive for production facilities that leads to multinational formation.
However, we did not discuss why the parent firm chooses to own the affiliate in that
location and operate as a single multinational firm. This is known as the internalization
motive.
As a substitute for horizontal FDI, a parent could license an independent firm to produce
and sell its products in a foreign location; as a substitute for vertical FDI, a parent
could contract with an independent firm to perform specific parts of the production
process in the foreign location with the best cost advantage. This substitute for vertical
FDI is known as foreign outsourcing (sometimes just referred to as outsourcing, where
the foreign location is implied).
Offshoring represents the relocation of parts of the production chain abroad and
groups together both foreign outsourcing and vertical FDI. Offshoring has increased dramatically
in the last decade and is one of the major drivers of the increased worldwide
trade in services (such as business and telecommunications services); in manufacturing,
trade in intermediate goods accounted for 40 percent of worldwide trade in 2008. When
the intermediate goods are produced within a multinational’s affiliate network, the shipments
of those intermediate goods are classified as intra-firm trade. Intra-firm trade represents
roughly one-third of worldwide trade and over 40 percent of U.S. trade.
What are the key elements that determine this internalization choice? Control over a
firm’s proprietary technology offers one clear advantage for internalization. Licensing another
firm to perform the entire production process in another location (as a substitute for
horizontal FDI) often involves a substantial risk of losing some proprietary technology. On
the other hand, there are no clear reasons why an independent firm should be able to replicate
that production process at a lower cost than the parent firm. This gives internalization
a strong advantage, so horizontal FDI is widely favored over the alternative of technology
licensing to replicate the production process.
The trade-off between outsourcing and vertical FDI is much less clear-cut. There are
many reasons why an independent firm could produce some parts of the production
process at lower cost than the parent firm (in the same location). First and foremost, an
independent firm can specialize in exactly that narrow part of the production process. As a
result, it can also benefit from economies of scale if it performs those processes for many
different parent firms.21 Other reasons stress the advantages of local ownership in the
alignment and monitoring of managerial incentives at the production facility.
But internalization also provides its own benefits when it comes to vertical integration
between a firm and its supplier of a critical input to production: This avoids (or at least
lessens) the potential for a costly renegotiation conflict after an initial agreement has been
reached. Such conflicts can arise regarding many specific attributes of the input that cannot
be specified in (or enforced by) a legal contract written at the time of the initial agreement.
This can lead to a holdup of production by either party. For example, the buying firm can
21Companies that provide outsourced goods and services have expanded their list of clients to such an extent that
they have now become large multinationals themselves. They specialize in providing a narrow set of services (or
parts of the production process), but replicate this many times over for client companies across the globe.
186 PART ONE International Trade Theory
claim that the quality of the part is not exactly as specified and demand a lower price. The
supplying firm can claim that some changes demanded by the buyer led to increased costs
and demand a higher price at delivery time.
Much progress has been made in recent research formalizing those trade-offs. This research
explains how this important internalization choice is made, by describing when a
firm chooses to integrate with its suppliers via vertical FDI and when it chooses an independent
contractual relationship with those suppliers abroad. Developing those theories is
beyond the scope of this textbook; ultimately, many of those theories boil down to different
trade-offs between production cost savings and the fixed cost of moving parts of the
production process abroad.
Describing which types of firms pick one offshoring option versus the other is sensitive
to the details of the modeling assumptions. Nonetheless, one robust prediction emerges
from those models when one compares either offshoring option to that of no offshoring
(not breaking up the production chain and moving parts of it abroad). Relative to no offshoring,
both vertical FDI and foreign outsourcing involve lower production costs combined
with a higher fixed cost. As we saw, this implies a scale cutoff for a firm to choose
either offshoring option. Thus, only the larger firms will choose either offshoring option
and import some of their intermediate inputs.
This sorting scheme for firms to import intermediate goods is similar to the one we
described for the firm’s export choice: Only a subset of relatively more productive (lowercost)
firms will choose to offshore (import intermediate goods) and export (reach foreign
customers)—because those are the firms that operate at sufficiently large scale to favor the
trade-off involving higher fixed costs and lower per-unit costs (production- or traderelated).
Empirically, are the firms that offshore and import intermediate goods the same set of
firms that also export? The answer is a resounding yes. For the United States in 2000,
92 percent of firms (weighed by employment) that imported intermediate goods also
exported. Those importers thus also shared the same characteristics as U.S. exporters:
They were substantially larger and more productive than the U.S. firms that did not engage
in international trade.
Consequences of Multinationals and Foreign Outsourcing
Earlier in this chapter, we mentioned that internal economies of scale, product differentiation,
and performance differences across firms combined to deliver some new channels for
the gains from trade: increased product variety, and higher industry performance as firms
move down their average cost curve and production is concentrated in the larger, more
productive firms. What are the consequences for welfare of the expansion in multinational
production and outsourcing?
We just saw how multinationals and firms that outsource take advantage of cost differentials
that favor moving production (or parts thereof) to particular locations. In essence,
this is very similar to the relocation of production that occurred across sectors when opening
to trade. As we saw in Chapters 3 through 6, the location of production then shifts to
take advantage of cost differences generated by comparative advantage.
We can therefore predict similar welfare consequences for the case of multinationals
and outsourcing: Relocating production to take advantage of cost differences leads to
overall gains from trade, but it is also likely to induce income distribution effects that leave
some people worse off. We discussed one potential long-run consequence of outsourcing
for income inequality in developed countries in Chapter 5.
Yet some of the most visible effects of multinationals and outsourcing occur in the
short run, as some firms expand employment while others reduce employment in response
CHAPTER 8 Firms in the Global Economy 187
to increased globalization. We mentioned in Chapter 4 that those employment changes due
to overseas plant relocations (along with plant closures due to import competition) account
for only a small fraction (2.5 percent) of all involuntary worker displacements in the
United States. Nevertheless, when such plant relocations do occur, they inevitably generate
some substantial costs for those affected workers. As we argued in Chapter 4, the best
policy response to this serious concern is still to provide an adequate safety net to unemployed
workers without discriminating based on the economic force that induced their
involuntary unemployment. Policies that impede firms’ abilities to relocate production and
take advantage of these cost differences may prevent these short-run costs for some, but
they also forestall the accumulation of long-run economy-wide gains.
No comments:
Post a Comment