The Bureau
of Economic Affairs (BEA) of the US Ministry of Commerce
has recently released preliminary results of the 1999
benchmark survey of US direct investment abroad. These
results, provided in the Survey of Current Business,
March 2002, enable an assessment of changes in the role
and impact of multinational corporations during the
years of globalization.
It was widely believed that economic liberalization
would lead MNCs to both expand their business operations
and relocate much of their existing operations to low-to-middle-income
countries in the Asia-Pacific region and in Latin America.
However, the BEA evidence is contrary to such expectations.
Between 1989 and 1999 there was little change in the
share of US MNCs in economic activity both in the US
and elsewhere. In fact, the share of US MNCs in world
GDP in 1999 (6.1 per cent) was slightly lower than what
it was in 1989.
Besides, although the shares of Asia-Pacific and Latin
America in the gross product of US majority-owned foreign
affiliates (MOFAs) worldwide went up from 15 per cent
and 9 per cent in 1989 to 18 per cent and 11 per cent
respectively in 1999, Europe still remains the most
important location for production by US MOFAs. The total
gross product of MOFAs in Europe in 1999 was US $321.6
billion out of a worldwide figure of US $ 561.2 billion,
with the United Kingdom and Germany accounting for over
half this gross product. In percentage terms, MOFAs
in Europe produced 57.31 per cent of the total gross
product of US MOFAs. Canada accounted for another US
$63.8 billion, or 11.37 per cent of the total gross
product of US MOFAs worldwide.
Chart
1 >> Chart
2 >>
The principal gainers of the rise in gross product of
MOFAs in the Asia-Pacific region between 1989 and 1999
were developed countries like Japan, Australia and Singapore.
Japan and Australia accounted for almost half of the
gross product of the MOFAs in the region and almost
9 per cent of total gross product of US MOFAs worldwide.
Together with Singapore and Taiwan province of China,
the total share of these economies in the gross product
of MOFAs in the region was about two-thirds.
Chart
3 >>
UK's share in the total gross product of MOFAs in 1999
was larger than that of Asia-Pacific, while Canada's
share, though significantly lower in 1999 than what
it used to be a decade ago, was still higher than the
share of all Latin American countries taken together.
Most countries where MFOAs accounted for a significant
share of the GDP in 1999 are in the developing world,
Nigeria, the Honduras and Malaysia being exceptions.
The share was the highest in Ireland (16.8 per cent),
followed by Singapore (10.7 per cent) and Canada (10
per cent).
All this points to the fact that, even though the periphery
has witnessed some increase in MNC activity, this growth
has not been at the expense of the presence of multinationals
in the metropolis. Even after including the MOFAs in
developed countries, the presence of US parents in the
global operations of MNCs remains strong. While in 1989
the share of US parents in the gross product of parents
and MOFAs put together was 76.6 per cent, a decade later
it was only marginally lower at 76.3 per cent. Indeed,
with trade restrictions easing, it has now become simpler
for MNCs to produce in one country and export the products
to others. So, while earlier MNCs had to often set up
production facilities in countries they intended to
sell in, there is no need for that any longer. The share
of export sales in total sales of US parents actually
increased from 6.71 per cent in 1989 to 8.05 per cent
five years later. If liberalization entails shifting
more and more of the production processes to the country
of final sale (or to countries near the country of final
sale) one would have expected a reduction in the share
of export sales in total sales of US parents. Though
this share came down to 7.10 per cent in 1999, it was
still higher than what it was in 1989. Thus, it is not
true that liberalization has always worked to the advantage
of developing countries so far as attracting investment
from MNCs are concerned. For many, it can work the other
way around, with MNCs relocating production facilities
away from some countries, and later, exporting to them.
If liberalization had indeed worked in favour of the
developing countries, one would have expected to see
more and more of the production being transferred to
the countries of final destination. The share of export
sales in total sales of US parents should have steadily
climbed down, a claim that the data fail to substantiate.
The table below gives an idea of the increase in assets
of MOFAs in different countries between 1989 and 1994
and between 1994 and 1999.
Table
1 >>
Even as the growth of total assets of MOFAs fell sharply
during 1994-99 as compared to the preceding five-year
period, the growth of assets of MOFAs in Canada during
1994-99 was more than thrice the growth during 1989-94.
The growth in Europe, although lower during 1994-99,
was still a respectable 32 per cent. Africa’s spectacular
performance in this regard may be attributed to the
low base period asset value, rather than to the region’s
emergence as a destination for new investment.
In terms of the number of employees, the share of US
parents recorded a fall, from 78.6 per cent in 1989
to 74.1 per cent in 1999. But this does not necessarily
imply that expansion of MOFAs is taking place faster
than expansion of their parents in the US. It may well
be true, and possibly is, that labour-intensive segments
of technology-intensive production processes are being
relocated to alternative sites in the developing world
where labour is cheap. Or, it may be the case that production
processes in developed countries are getting increasingly
mechanized, thereby not allowing the employed labour
force to grow as fast as it would otherwise have. In
fact, the gross products of US parents and their affiliates
grew at about the same rate in 1999. However, while
employment in the affiliates increased 4 per cent, parent
employment declined.
It must be noted that the view that lower wages in some
developing countries makes them an attractive site for
multinational investment is not borne out by the evidence
relating to US MNCs. A sample survey carried out in
1992 in thirteen high-wage and fourteen low-wage economies(the
latter included economies like Hong Kong, Singapore
and Taiwan province of China) revealed that 65 per cent
of employment by manufacturing MOFAs was in relatively
high-wage countries. If one takes into consideration
the same two sets of countries and calculates the percentage
of employment by manufacturing MOFAs in 1999, it is
found that there has been only a 2 per cent decline
in the share of the said employment in the thirteen
high-income countries.
The share of US parents in the worldwide gross product
of US MNCs has remained near-constant throughout the
1990s in almost all sectors, including manufacturing,
finance, insurance and real estate. Only the services
sector has seen a significant decline in this share
during the second half of the 1990s. This sector has
seen a rise in its share in gross product of all industries,
mainly owing to the growth of computer and data-processing
services. And a larger part of this growth has taken
place in the periphery, as parents have been outsourcing
many of these activities to take advantage of the cheap
but skilled work force that English-speaking developing
countries provide. In 1989, out of the total gross product
of $67 billion of US MNCs worldwide in the services
sector, US parents accounted for US $57.1 billion, or
about 85 per cent. In 1999 their share was US $178 billion
out of a total of 220.8 billion, or about 81 per cent.
US MNCs are the leading spenders on research and development
(R&D). However, R&D is still carried out mainly
by US parents: their share in R&D expenditure has
increased during the 1990s, from 83 per cent in 1989
to 87 per cent a decade later. In 1999, US parents accounted
for 68 per cent of total R&D expenditure in the
US. This reflects the fact that large multinational
firms still use technology as a means to market leadership,
and points to control that patenting ensures as a means
to higher profits. The fact that parent firms account
for a large share of R&D expenditures even within
US MNCs shows that the tendency of firms to locate research
activities at or near their headquarters still persists,
reducing the possibility of scientific and technological
spin-offs from MNC investment in the periphery. Many
MOFAs do not undertake R&D at all. While US parents
undertaking R&D accounted for 61 per cent of the
gross product of all US parents in 1999, the share of
MOFAs undertaking R&D in the gross product of all
MOFAs in the same year stood at a mere 35 per cent.
The ratio of R&D expenditures to the gross product
(R&D intensity) of all US parents in 1999 was 6.8
per cent. In contrast, the corresponding figure for
MOFAs was only 3.3 per cent. In terms of R&D intensities,
MOFAs in Israel and Sweden had very high percentages
of 21.3 and 15.6, respectively; the R&D intensity
in China was 7.8 per cent, followed by the OECD countries.
In 1999, MOFAs in developing countries, leaving out
China and Brazil (with R&D intensity of 1.9 per
cent), had an abysmally low average R&D intensity
of 1.3 per cent.
The ratio of R&D expenditures to the gross product
of R&D-undertaking MOFAs in 1999 was 9.4 per cent,
while that of R&D-undertaking parents was 11.3 per
cent. For Israel it was 50.8 per cent, for Sweden 41.4
per cent, and for China 22.5 per cent. At the other
end of the spectrum, the ratio of R&D expenditures
to the gross product of R&D-undertaking MOFAs in
Brazil in the same year was only 4.3 per cent, and in
other developing countries the average was 6 per cent.
In the communications equipment segment, however MOFAs
spent more on R&D in 1999 than their US parents.
Among R&D-undertaking communications equipment firms,
MOFAs had an R&D intensity of 50 per cent, while
that of US parents was only 38 per cent. The figures
are exactly the opposite in computers and peripheral
equipment, with MOFAs in the field having an R&D
intensity of only 8 per cent while parent firms had
27 per cent.
Manufacturing parents have been the largest spenders
on R&D with R&D intensities being particularly
high in computers and electronic products (especially
communications equipment), chemicals (especially pharmaceuticals
and medicines), and transportation equipment. Almost
all US parents manufacturing computers and electronic
products undertake R&D: while their R&D intensity
in this industry in 1999 was 30 per cent, the industry
figure for all parents in computers and electronic products
in the same year stood at 29 per cent. Outside manufacturing,
parents in publishing, computer systems design and related
services also exhibited relatively high R&D intensities.
Of a total of US $18.4 billion spent by MOFAs on
R&D in 1999, US $15.7 billion were spent in research
centres in developed countries, with those in the
UK, Germany, Canada, Japan and France accounting for
around US $10.8 billion of the expenditures. The UK
and Germany together accounted for more than two-fifths
of all R&D spending by MOFAs.
Chart
4 >>
The share of R&D-undertaking MOFAs in the gross
product of all MOFAs has been higher in most of the
developed countries compared to the developing world.
For all countries taken together, this share was 34.9
per cent in 1999. In Germany it was 49 per cent, in
Singapore 48.4 per cent, in France 43.7 per cent,
in the UK 42.7 per cent and in Israel 41.9 per cent.
Among the developing countries, the share of R&D-undertaking
MOFAs in the gross product of all MOFAs has been particularly
high in Brazil, standing at 43.8 per cent in 1999.
If one leaves out Brazil and China (34.5 per cent),
this share was only 22.3 per cent for developing countries
in that year.
The figures above imply that not many of the MOFAs
in developing countries engage in R&D, and that
those who do have not contributed significantly to
the gross product. It may well be the case that MOFAs
in developing countries are more interested in exploiting
the natural resources of those countries, oil and
natural gas in particular. MOFAs in the manufacturing
sector, which account for most of the R&D expenditure
made by MOFAs, are mostly located in the OECD countries.
Finally, if we look at the expansion of US MNCs in
1999, we find that more than half of the new affiliates
of US MNCs have been acquired and only a little more
than 46 per cent of the new affiliates have been newly
established, resulting in investment in green-field
projects. This means that only in a little more than
half the cases does the creation of new affiliates
involve investment in new capacity or the creation
of new employment opportunities in the host countries.
On the contrary, take-overs by US MNCs have often
led to retrenchment of workers in these companies.
Europe is still the most popular location for new
affiliates. In 1999, 56 per cent of all new affiliates
were accounted for by Europe. The new European affiliates
produced 68 per cent of the gross product of all new
affiliates put together, and accounted for 61 per
cent of all the people employed by the new affiliates.
Canada accounted for another 8.5 per cent, and only
18 of the 1,077 new affiliates were set up or acquired
in Africa.
USDIA
US direct investment abroad (USDIA), on a historical
cost basis stood at US $1244.7 billion at the end
of 2000. However, hardly any money has been invested
till date in poor countries, particularly those in
Africa: the share of USDIA that accrued to Africa
till 2000 was only 1.27 per cent of total USDIA worldwide.
US $233.4 billion or about 19 per cent of USDIA had
gone to the United Kingdom; another 10 per cent had
gone to Canada; and then Netherlands had a 9 per cent
share. Within Europe, USDIA on a historical cost basis
was only US $672 million in Greece in the year 2000.
In Central America, it amounted to only US $115 million
in the Honduras and US $904 million in Guatemala;
Ecuador, in South America, has received only US $838
million till 2000.
Chart
5 >> Chart
6 >>
Of the US $113.9 billion worth of fresh USDIA made
worldwide in 2000, Europe got US $60.4 billion and
Canada US $15.4 billion, together bagging two-thirds
of the investment. Africa got less than US $1 billion,
with countries like Nigeria and South Africa witnessing
an outflow of US money. South America attracted close
to US $5 billion worth of USDIA in 2000, with money
actually flowing out of Ecuador. The same was true
of the Honduras in Central America.
The shares of individual countries in the total investment
of US $60.4 billion that went to Europe were as follows.
The UK got more than a third, and the Netherlands,
Ireland, Switzerland and Italy together got half.
That is, these five countries garnered US $50.3 billion,
with the rest of Europe getting only US $10.1 billion.
Many European nations also saw their net USDIA dwindling
during 2000, Belgium, France, Austria and Finland
being the prominent ones among them.
The table below gives the shares of the major destinations
of USDIA on a historical cost basis in 2000 and the
shares of fresh USDIA made in 2000. One can see that
only in Europe and Canada (besides the Middle East,
whose share of USDIA is very low) are the shares of
fresh USDIA in 2000 higher than the shares on a historical
cost basis.
Table
2 >>
These figures reject the widespread belief that increasing
shares of fresh USDIA are going to the developing
countries, and negate all claims about the erstwhile
East European countries becoming a hotbed for investment
by US MNCs in recent years. USDIA in Eastern Europe
at historical cost in 2000 was only US $11 billion,
with new net investment in the year 2000 being a meagre
US $1.4 billion. Further, most of this new investment
was in the financial sector and did not contribute
much to the growth of infrastructure or manufacturing
industries in the region. While Asia and the Pacific
region saw a US $17.7 billion rise in USDIA in 2000,
Japan got US $6.2 billion, and Hong Kong and Singapore
together got another US $6.3 billion.
China, an emerging Asian market, got only US $1.6
billion during the year. This is contrary to the perception
that, since it is the largest recipient of FDI among
developing countries, China must be attracting a lot
of USDIA as well. The share of China in USDIA on a
historical cost basis till 2000 was only 0.77 per
cent of total USDIA, and 4.80 per cent of USDIA in
the Asia-Pacific region. Out of the fresh USDIA made
in 2000, China got a mere 1.33 per cent. Its share
of the fresh USDIA made in the Asia-Pacific region
in 2000, although standing at a not-too-significant
8.57 per cent, reveals that China’s share in USDIA
is rising and that its importance as a destination
for investment by US MNCs is increasing.
Much of the increase in USDIA in 2000 was accounted
for by reinvested earnings of affiliates except in
the UK, Switzerland and Italy, where the increases
mainly reflected acquisition. In the UK, the acquired
firms were mostly existing British businesses, while
in Switzerland and Italy existing foreign businesses
were acquired. So there has been little additional
capacity created through fresh investment by US parents
in foreign countries in 2000, not even in Europe.
Acquisitions rarely create additional capacity, and
are often followed by downsizing of the existing work
force rather than expansion of acquired firms.
Most of the USDIA that Africa and the Middle East
have received till date has gone into exploiting petroleum
resources in these regions. As we have already observed,
in the section on R&D, MOFAs do not spend much
in these regions. USDIA figures on a historical cost
basis for 2000 for the Middle East show that while
about US $2.9 billion was invested to explore petroleum,
manufacturing industries got only US $2.5 billion.
The figures for Africa reveal even greater exploitation
of the region’s oil and natural gas resources by US
MNCs. USDIA in Africa in 2000 on a historical cost
basis was around US $10 billion in the petroleum sector,
but only a meagre US $2.2 billion in manufacturing.
Of the USDIA in manufacturing in Africa, a little
less than US $950 million went to South Africa.
On looking at recent direct investment by the US in
different countries and comparing its rise in 2000
over what it was in 1999, one finds that USDIA has
grown by only 6 per cent in Africa and about 8.5 per
cent in Latin America. In contrast, USDIA in Canada
and Europe has grown at 14 per cent and 10 per cent
respectively. In the Middle East it registered a 12.7
per cent, this was probably owing to the fact that
the assets of MOFAs in the region had fallen sharply
in the five years preceding 1999. Overall, USDIA rose
10 per cent during the year.
Also, the distribution of whatever little USDIA has
gone to the developing countries reflects an extremely
skewed pattern, with a few countries getting almost
the entire USDIA. For example, USDIA on a historical
cost basis in 2000 in South America stood at US $79,354
million, of which US $35,560 million, or about 45
per cent went to Brazil, 18 per cent to Argentina,
14 per cent to Chile, and 11 per cent to Venezuela.
The distribution was even more uneven in Central America.
Of the US $74,754 million that the region attracted
as USDIA till 2000, Mexico and Panama got more than
47 per cent each. Out of the USDIA going to other
countries in the western hemisphere, Bermuda got more
than 63 per cent and the Caribbean got 24 per cent.
The investment in Bermuda does not really help the
country much, as it is used as an offshore tax-haven
by US parents, not as a destination for setting up
production facilities.
These data relating to US MNCs, thus, fly in the face
of the optimism in the developing world about increases
in FDI after liberalization. While much of the expansion
and investment by US MNCs is still being directed
at the OECD countries, the increased spate of acquisitions
and mergers would in all probability increase the
dominance of a few MNCs over the global economy. That
increase in dominance falls far short of the expected
increase in new production facilities and employment
opportunities on which the case for economic liberalization
rests.
September 7, 2002.
|