From the early internationalisation of corporate
technology to global technology sourcing


by John Cantwell

 

Professor John Cantwell
Department of Economics
University of Reading
PO Box 218, Whiteknights
Reading RG6 6AA
UK

telephone: +44 (0) 118 9875123
fax: +44 (0) 118 9750236

email: J.A.Cantwell@rdg.ac.uk

 

 

From the early internationalisation of corporate
technology to global technology sourcing

 

Abstract

The early work of both Dunning and Vernon laid the crucial foundations for the large body of subsequent research on technological change and TNCs. While corporate technology creation in TNCs is often more widely internationally dispersed than Vernon had originally supposed, the notion of the locational agglomeration of innovation in key centres remains highly relevant. The TNC now provides a means of linking alternative streams of innovation in geographically separate centres through the organisation of the exchange of knowledge across national boundaries, as emphasised by Dunning in his latest work. This shift in the role of TNCs, from exploiting home-base technology in international markets to the international sourcing of technology, is illustrated here by revisiting the evidence on technologically active US-owned TNCs in the UK and Europe, the case with which Dunning and Vernon had begun. It is shown that US TNCs developing technology locally in the UK have moved away from their historical focus on the industries in which they were strongest at home (notably in electrical equipment) towards industries in which indigenous UK companies have the greatest technological expertise (notably in chemicals and pharmaceuticals).

Introduction: the early expansion of US TNCs in Europe

In the earliest version of the product cycle model, Vernon (1966) envisaged that transnational corporations (TNCs) concentrate their technological development in their home country. The three theoretical justifications for this view were that there are scale economies in the research and development (R&D) function, locational economies of agglomeration in new product development, and a stimulus to innovation from the demand of high income consumers and skill-intensive downstream production facilities, in the 1960s especially for US-owned TNCs in the advanced US economy (see Cantwell, 1995). However, Vernon based this aspect of the PCM on the evidence mainly of the early post-war experience of US TNCs, as represented by the findings of the Harvard Multinational Enterprise Database project (Vaupel and Curhan, 1973).

With the benefit now of a much wider range of evidence on the extent of the internationalisation of R&D in large TNCs, it has become clear that technological development is not always highly centralised in the parent company in the home country (Lall, 1979, Mansfield, Teece and Romeo, 1979, Casson, 1991, Granstrand, Håkanson and Sjölander, 1992, Pearce, 1997), even if in general the home country is the most important single location for R&D (Patel and Pavitt, 1991, Patel, 1995). US TNCs had carried out significant levels of technological development in their affiliates located in the UK and Europe in the interwar period just as they are doing so again today, while UK- Swiss- and Dutch-owned TNCs had already highly internationalised their R&D by the 1960s (Cantwell, 1995). Thus, the extreme case of a very high degree of concentration of R&D in the home country of the TNC is a special case. It is a special case that tends to be representative of countries whose outward investment in other industrialised countries grows out of technological hegemony or an outstanding record in domestic innovation - such as the US in the early post-war years, and Japan since the 1970s.

Most notably in the electrical equipment industry, US TNCs had a history in the interwar years of developing some new technologies in the UK and Europe. Although they did so only to a lesser extent after the war, Dunning (1958, 1970) observed how their local development efforts spilled over to help improve productivity and export performance in some industries in the UK, and subsequently more widely in Europe. He postulated that while occasionally the local presence of US-owned R&D might drive out weaker indigenous players, that in general it would tend to enhance the technological capability of the host country. Following up on this line of argument Cantwell (1989) found that US inward direct investment in Europe had the most favourable dynamic effect upon the technological efforts of indignous firms in industries in which local companies had a strong technological tradition of their own, which was revived by the competitive stimulus provided by innovative US TNC new entry.

However, in the early stages of the internationalisation of technological development in TNCs historically the primary motive of the firms was to adapt products to local tastes and living and working conditions, local production and regulatory requirements, and local resource availabilities or scarcities (Cantwell, 1995). Hence, in terms of broadly defined industrial groups, US-owned R&D activity in the UK initially reflected mainly the existing technological strengths or competitive advantages of the relevant parent companies, the focus on electrical equipment being a result of the interwar technological preeminence of companies such as General Electric, AT&T and ITT. Even today, much R&D in foreign-owned affiliates is still largely aimed at adapting technologies to local markets and environments (Pearce and Singh, 1992, Fors, 1998). As observed by Dunning (1958, 1998), US-owned affiliates in Britain first engaged in local research mainly to adapt technology to British production conditions, reflecting principally the structure of technological activity in the US economy. Yet even as early as the 1950s, he notes how some US-owned affiliates in the UK tapped in to local sources of expertise and created technologies that proved useful to their US parent companies, so the typical direction of transfer was reversed. More recently, TNCs have increasingly developed a more globalised approach to innovation that embodies more decentralised but better integrated R&D activities to gain access to the main centres of R&D excellence for their industry (Dunning, 1994, Cantwell, 1995, Pearce, 1997, Florida, 1997, Cantwell and Janne, 1999). Over time, TNCs increasingly and gradually take advantage of host locations as centres for innovation and thus favour sectors in which the host country has a relative competitive advantage. Therefore, we would expect that the technological development of many US-owned affiliates has become more sophisticated and specialised to exploit British-specific technological expertise and skills (as might have been forecast from some suggestions of Dunning, 1958) as a source of complementary technology for the company of which they are part as a whole.

This paper revisits the early perspectives on technological change in TNCs that derive from the work of Vernon and Dunning, with particular reference to the case of the activities of US TNCs in the UK and Europe, on which their earlier studies had concentrated. More particularly it compares the extent, evolution and industrial patterns of research undertaken by US TNCs in the UK and Europe in each of three phases - in the interwar period, on which Vernon and Dunning had only limited evidence, in the early post-war period which they documented extensively, and in the more recent period of take-off in internationally integrated TNC strategies. The characteristics of these three phases in the technological accumulation of the largest TNCs have been discussed further in Cantwell and Piscitello (1999b). Our evidence relies on a database of corporate patents granted in the US to the world's largest industrial firms, which allows us to distinguish the location of the research facility originally responsible for each patented invention, as well as its ownership. This enables us to monitor the technological development efforts of US-owned affiliates located in Europe, to examine their industrial structure and how it has changed over time.

Data

The technological activity of US TNCs in Europe is examined using data on patents granted in the United States to the largest US-owned and European-owned firms in the historical periods 1920-39 and 1940-68, and then in the more recent period 1969-95. A large literature has pointed out the advantages as well as the limitations of patent statistics as an internationally comparable indicator of technological activity (e.g. Pavitt, 1988, Acs and Audretsch, 1989, Griliches, 1990). As mentioned already, our patent database distinguishes both the country of origin of the invention (or the location of the corporate research facilities responsible) and the country of location of the parent firm. All patents granted under the names of affiliates have been consolidated into the relevant corporate group, but the historical dataset for the first two periods relies on a smaller set of companies. There are 284 corporate groups in the historical case (of which 92 are US-owned and 192 are European-owned, including 65 UK-owned firms), and 603 corporate groups for the period since 1969 (of which 336 are US-owned 267 are European-owned, of which 84 are UK-owned companies). The patenting of large firms from 1920-68 was recorded manually from the US Index of Patents and the US Patent Gazette, while from 1969 onwards equivalent information has been computerised by the US Patent Office. The firms selected for the historical patent search were identified in one of three ways. The first group consisted of those firms which have accounted for the highest levels of US patenting after 1969; the second group comprised other US, German or British firms which were historically among the largest 200 industrial corporations in each of these countries (derived from lists in Chandler, 1990); and the third group was made up of other companies which featured prominently in the US patent records of earlier years (a method that proved most significant for a number of French firms that had not been identified from other sources). For a further discussion of the historical data see Cantwell (1995).

For 1969-95, the corporate groups are drawn from the world's largest 792 industrial companies (as derived from Fortune, and listed in Dunning and Pearce 1985, of which 730 had recorded patenting activity for the 1969-95 period), with the addition of 54 technologically large companies apparently missed from the Fortune listings to make 784 corporate groups in all. Of these, 181 of the largest firms are of neither US nor European origin, and so they are not considered for our current purposes. The consolidated firms are also allocated to their primary industry of output according to the product distribution of their sales, so that corporate patenting was divided into ten broad industrial groups. European countries are defined as Germany, UK, Italy, France, Netherlands, Belgium and Luxembourg, Switzerland, Sweden, Denmark, Ireland, Spain, Portugal, Greece, Austria, Norway and Finland. The problem of the variation in the propensity to patent the results of innovation over time, amongst industries, technological sectors and nations is avoided by constructing measures from the US patent statistics in the form of shares and ratios rather than absolute numbers. Some figures should however be interpreted with care when the numbers of patents registered is low. Consequently slight changes may be reflected in large percentage increases or decreases in the numbers presented.

Countries’ industrial patterns of technological specialisation can be observed by means of a ‘Revealed Technological Advantage’ index (RTA) as developed by Soete (1987), Cantwell (1989), and Patel and Pavitt (1991). The RTA index of the US and the UK as hosts to the research of the largest firms can be calculated across industrial groups of companies, and is defined as that country’s share of all US patenting in a given industry relative to its share of all US patenting in all industries - all large firms patenting in the US, irrespective of their country of ownership or of where technological development is located. Denoting by Pij the number of US patents of the host country j in a particular industry i, the RTA index for each country in that industry is defined as (Pij / Sj Pij) / (Si Pij / Sij Pij). The index varies around unity, so a value greater than one suggests that the host country is comparatively advantaged or specialised in the considered industry in relation to other countries, and a value less than one shows comparative disadvantage. Similarly, the RTA index is defined across industries for groups of US and British firms at home respectively, US firms located in the UK and UK firms located in the US.

 

The early view from the UK host country and US home country perspectives

The historical dominance of the electrical equipment industry in the European-located technological development of US TNCs is illustrated in Table 1, a dominant position that reflected the preeminence of US companies in this field, as remarked upon above. As measured by corporate patenting, roughly two-thirds of the R&D conducted by US-owned affiliates in the UK and in Europe in the interwar and early post-war years was organised and managed by firms in the electrical equipment industry. In the inter-war period US TNCs in non-electrical machinery had also carried out a substantial local technological effort especially in the UK, a sign of strength that also reflected excellence at home. Yet after the war, US-owned affiliates were rather more active than they had been previously in chemicals and pharmaceuticals, motor vehicles and scientific instruments, but less so in non-electrical machinery.

Table 1: Industry shares of US patents of the largest US firms attributable
to research located in the UK and Europe, 1920-68.

. .

US affiliates located

.

US affiliates located

.
. .

in the UK

.

in Europe

.

.

.

1920-39

1940-68

1920-39

1940-68

1

Food, drink and tobacco

0.00

0.61

0.00

0.66

2

Chemicals and pharmaceuticals

1.27

7.95

3.42

13.40

3

Non-electrical machinery

30.35

13.64

15.12

8.54

4

Electrical equipment and computing

65.10

63.86

75.33

63.99

5

Motor vehicles

2.38

5.73

2.88

4.87

6

Rubber and plastic products

0.05

0.35

0.06

0.35

7

Non-metallic mineral products

0.21

0.35

0.27

0.64

8

Coal and petroleum products

0.00

2.86

2.36

3.22

9

Professional and scientific instruments

0.63

4.47

0.56

3.57

10

Other manufacturing

0.00

0.18

0.00

0.76

. .
.
.
.
.
.

Total all industries

100.00

100.00

100.00

100.00


Source: US patent database compiled by John Cantwell at the University of Reading,
with the assistance of the US Patent and Trademark Office.

The fall in the technological contribution of US-owned affiliates in the UK and Europe after the war can be viewed either from the host country or the home country perspective. On the UK host country side, the participation of US TNCs in the research of large firms located in Britain dropped from 32.8% to 19.5%, and it fell back especially in the electrical equipment, non-electrical machinery and motor vehicle industries, as shown in Table 2. However, although US TNCs in their areas of greatest technological strength refocused their innovative efforts at home in the US (which was then reflected, as we have seen already, in Vernon's earliest version of the product cycle model), occasionally in other industries there was a more moderate expansion of US-owned research abroad even in this early post-war period. In particular in the UK, there was an increase from a low base in the US-owned share of development in chemicals and pharmaceuticals, in which US companies had improved their competitiveness vis-à-vis European firms by comparison with the interwar years, and a dramatic extension in the US control over UK-located R&D in scientific instruments (most notably in photographic equipment). This latter involvement can be traced back to the establishment of Eastman-Kodak's R&D facility in Harrow, near London in 1928 (Wilkins, 1974). Despite the relative weakness of indigenous British company effort in this sector, Kodak built a European research base from this investment which later drew in the R&D of other foreign-owned firms, an essentially foreign-owned TNC technology clustering of the kind observed in the Singaporean experience today.

Table 2: Share of patenting activity from UK-located large firm research
attributable to US-owned companies, by industry of the parent firm, 1920-68 (%).

.

.

1920-39

1940-68

1
Food, drink and tobacco

0.00

4.40

2
Chemicals and pharmaceuticals

2.43

7.14

3
Non-electrical machinery

54.30

31.62

4
Electrical equipment and computing

64.38

33.41

5
Motor vehicles

40.91

13.72

6
Rubber and plastic products

0.44

2.52

7
Non-metallic mineral products

4.44

6.49

8
Coal and petroleum products

0.00

12.96

9
Professional and scientific instruments

18.46

75.37

10
Other manufacturing

0.00

0.27

.
Total all industries

32.80

19.49


Source: As for Table 1.

From the US home country angle, in the interwar period the equivalent of over one-tenth of the very substantial level of research in electrical equipment carried out at home by large US firms was conducted in their European-located affiliates. About one-third of this was done in the UK (see Table 3). An even higher relative share was recorded by UK-located affiliates in non-electrical machinery, perhaps owing to Britain's continuing attractiveness for the development and exploitation of the basic mechanical technologies in which her traditional strength lay, rather than for research in the newer science-based industries, in which Germany had forged ahead. After the war, the decline in internationalisation of R&D in US-owned TNCs is clear. In comparison with the domestic research of the same large US companies, their foreign-located technological development fell from 6.6% to 3.1% in Europe, and from 2.6% to 1.5% in the UK. Even in scientific instruments in which their was some positive dispersion of R&D, the early post-war share reached no higher than 4.0% in Europe as a whole, and 2.4% in the UK.

Table 3: Patenting of US-owned companies from their European-located facilities as a proportion of patenting from parent company facilities in the US, in the UK and in Europe as a whole, 1920-68.

.. ..

Host locations

.
.
.
. .

Europe

.

UK

.
. .

1920-39

1940-68

1920-39

1940-68

1 Food, drink and tobacco

0.00

1.61

0.00

0.70

2 Chemicals and pharmaceuticals

1.98

1.92

0.29

0.54

3 Non-electrical machinery

6.18

2.82

4.86

2.13

4 Electrical equipment and computing

10.73

5.94

3.63

2.81

5 Motor vehicles

1.84

1.35

0.60

0.75

6 Rubber and plastic products

0.13

0.39

0.04

0.19

7 Non-metallic mineral products

0.77

1.09

0.24

0.28

8 Coal and petroleum products

2.89

0.81

0.00

0.34

9 Professional and scientific instruments

1.32

4.00

0.59

2.37

10 Other manufacturing

0.00

0.53

0.00

0.06

. Total all industries

6.58

3.07

2.58

1.45

Source: As for Table 1.

 

Patterns of international technological specialisation

Viewed from the historical perspective, Vernon's (1966) model remains useful in its association of the pattern of technological leadership through innovation in the home country and the industrial structure of outward direct investment. Using the RTA index derived from the data on corporate patenting as defined above, an index value greater than unity represents a position of relative innovative strength, or 'revealed technological advantage'. The cross-industry RTA distributions show that the profile of technological specialisation of US-owned affiliates in the UK roughly matched that of US companies at home, in electrical equipment and non-electrical machinery in the interwar years, and in scientific instruments after the war. Only in motor vehicles did the strength of US TNCs at home not lead to much local technological effort in their UK-located affiliates, as can be seen from Table 4.

 

Table 4: Revealed technological advantages across industries of US and British firms, 1920-68.

. .

US affiliates located

.

UK as a host

.

US firms in the US

.
. .

in the UK

.

country

.
.
.

.

.

1920-39

1940-68

1920-39

1940-68

1920-39

1940-68

1
Food, drink and tobacco

0.00

0.52

0.58

2.32

1.30

1.08

2
Chemicals and pharmaceuticals

0.07

0.34

1.00

0.92

0.67

0.90

3
Non-electrical machinery

1.93

1.41

1.17

0.87

1.02

0.96

4
Electrical equipment and computing

1.53

1.98

0.78

1.16

1.08

1.03

5
Motor vehicles

0.28

0.57

0.22

0.82

1.20

1.11

6
Rubber and plastic products

0.02

0.14

1.43

1.12

1.15

1.13

7
Non-metallic mineral products

0.11

0.22

0.83

0.67

1.21

1.14

8
Coal and petroleum products

0.00

0.24

0.11

0.36

1.11

1.02

9
Professional and scientific instruments

0.23

1.82

0.41

0.47

1.02

1.12

10
Other manufacturing

0.00

0.04

6.85

2.60

0.51

0.91


Source: As for Table 1.

In other words, in these early years US TNCs used their foreign affiliates mainly to exploit and extend a technological advantage developed at home, and so as to better serve the relevant foreign markets (Cantwell, 1995, Cantwell and Piscitello, 1999b). Or in Kuemmerle's (1996) teminology, their European-located research was primarily of a home-base exploiting kind, and not home-base augmenting. Far from attracting 'local technology sourcing' or what Dunning (1995) has more recently described as 'asset-seeking investment' by US-owned TNCs in the fields of local British technological strength, the interwar UK specialisation in chemicals and pharmaceuticals was essentially attributable to indigenous firms (notably ICI and the synthetic fibre companies, British Celanese and Courtaulds - see Cantwell and Barrera, 1998). At that time Du Pont had an alliance with ICI for joint technology development, under which cross-licensing agreement they ensured a geographical separation of their markets. Conversely, in the sectors of greater US TNC strength, the UK's technological specialisation in non-electrical machinery before the war, and in electrical equipment in the years after the war, owed much to the US presence and local extension of technologies that US TNCs had pioneered at home. As shown in Table 2, US-owned firms accounted for over 50% of domestic development efforts in the former case, and for over 30% in the latter case. Indeed, the early post-war revival of British research in electrical equipment might be traced to the dominant role of the US corporate giants in establishing a local UK technology base in the interwar period, when they were responsible for nearly 65% of locally generated patents in this industry.

The transformation towards international technology sourcing in the modern era of globalisation

In recent years the historical position has been reversed, in that US-owned affiliates in the UK have switched out of technological development in the electrical equipment industry in favour of chemicals and pharmaceuticals, as shown in Table 5. The US foreign participation in British-located research has concentrated increasingly in the industries of chemicals and pharmaceuticals (from 26.7% in 1969-77 to 39.6% in 1987-95), and professional and scientific instruments (from 3.9% to 6.9% over the same period), resulting in higher shares for the UK than for other European locations. US shares have also increased in the food, drink and tobacco and non-metallic and mineral products sectors. In contrast between 1969 and 1995, the proportion of US affiliate research in the electrical equipment and computing industry fell the most in the UK from 34.9% to 23.1%, while the research share of US-owned affiliates in that industry was relatively less important overall in the UK than in other European countries.

Table 5: Industry shares of US patents of the largest US firms attributable
to research located in the UK and Europe, 1969-95.

 
US affiliates located in the UK
.
.
US affiliates located in Europe
.
.

.

1969-77
1978-86
1987-95
1969-77
1978-86
1987-95
1
Food, drink and tobacco
1.93
2.28
2.18
1.51
1.78
2.20
2
Chemicals and pharmaceuticals
26.70
33.74
39.57
24.76
25.28
29.05
3
Non-electrical machinery
11.56
13.43
11.57
10.70
11.56
10.11
4
Electrical equipment and computing
34.91
31.03
23.15
39.15
37.52
31.46
5
Motor vehicles
5.66
5.30
7.07
4.24
4.53
6.70
6
Rubber and plastic products
2.07
0.86
0.08
2.64
1.77
1.23
7
Non-metallic mineral products
2.24
2.28
2.75
1.93
3.14
4.67
8
Coal and petroleum products
3.10
3.18
3.40
4.35
4.28
3.35
9
Professional and scientific instruments
3.86
3.10
6.88
4.65
2.85
5.40
10
Other manufacturing
7.97
4.79
3.36
6.07
7.28
5.84
.
Total all industries
100.00
100.00
100.00
100.00
100.00
100.00

Source: Cantwell, Dunning and Janne (1999).


From the UK host country perspective, the participation of US-owned affiliates has not recovered its interwar share of 32.8% (Table 2), but it has risen gradually from the early post-war figure of 19.5% to 20.5% in 1969-77, and back to 25.7% in 1987-95, as indicated in Table 6. Once again, the steady increase in the influence of US TNCs within the UK's corporate R&D owes especially to an increased presence in chemicals and pharmaceuticals, in which the US-owned share climbed from 2.4% in the interwar period and 7.1% in the early post-war years (Table 2) through 18.0% in 1969-77 to as much as 28.9% in 1987-95. Over the same period the equivalent share of US TNCs in the UK electrical equipment industry fell from the height of 64.4% in the interwar period to 33.4% in 1940-68 (Table 2), holding at 32.9% in 1969-77 before slipping back further to 23.4% in 1987-95. In other sectors the contribution of US TNCs to British corporate R&D recovered somewhat between 1969 and 1995 in non-electrical machinery and motor vehicles (Table 6), but not back to the levels they had attained in 1920-39 (Table 2); while the tremendous post-war expansion of US-owned R&D in the UK in professional and scientific instruments led already to a peak share in this industry of over 90% by 1969-77, and it has not fallen below 90% since, as the local technological agglomeration of US TNCs operating in the UK has been preserved in this field.

Table 6: Share of patenting activity from UK-located large firm research attributable
to US-owned companies, by industry of the parent firm, 1969-95 (%).

.

.

1969-77

1978-86

1987-95

1

Food, drink and tobacco

11.36

10.84

13.51

2

Chemicals and pharmaceuticals

17.99

23.20

28.91

3

Non-electrical machinery

21.74

26.91

37.97

4

Electrical equipment and computing

32.92

33.39

23.39

5

Motor vehicles

8.56

11.49

22.48

6

Rubber and plastic products

13.48

11.34

2.82

7

Non-metallic mineral products

19.29

25.44

48.32

8

Coal and petroleum products

11.81

14.59

14.78

9

Professional and scientific instruments

91.80

90.80

90.91

10

Other manufacturing

20.16

12.82

9.30

.

Total all industries

20.51

23.01

25.71


Source: As for Table 5.

Likewise from the US home country viewpoint, the foreign share of US TNC R&D has remained below its interwar peak. Despite increasing from 4.3% in 1969-77 to 5.6% in 1987-95 in Europe as a whole, the share is still below the 6.6% of 1920-39, while the rise in the UK-located share of US TNC technological development from 1.5% in 1969-77 to 1.7% in 1987-95 leaves it well below its interwar 2.6% (see Table 7, in comparison with Table 3). However, once again the increases in US-owned local R&D are particularly noticeable in chemicals and pharmaceuticals, from 4.5% in 1969-77 to 7.3% in Europe as a whole, and from 1.8% to 3.1% in the UK (Table 7), both of which are well above the comparable interwar shares of activity by US TNCs outside the US (given in Table 3). With respect to technological activity located in the UK, as emphasised already from Table 5, the most notable feature is that the switch towards an increasing reliance on a UK location in chemicals and pharmaceuticals contrasts sharply with the move away from the historical internationalisation of US-owned electrical equipment companies conducting research in the UK (Cantwell 1995). From a UK located share of US TNC R&D in the electrical equipment industry in the US of 3.6% in 1920-39 (Table 3), it had fallen to 1.9% in 1969-77, and from there to as little as 1.2% in 1987-95 (Table 7).

Table 7: Patenting of US-owned companies from their European-located facilities as a
proportion of patenting from parent company facilities in the US, in the UK and in Europe as a whole, 1969-95.

. .

Host locations

.
.
.
.
.
. .

Europe

.
.

UK

.
.
. .

1969-77

1978-86

1987-95

1969-77

1978-86

1987-95

1

Food, drink and tobacco

2.77

4.95

6.69

1.27

2.05

2.04

2

Chemicals and pharmaceuticals

4.52

5.88

7.27

1.75

2.54

3.06

3

Non-electrical machinery

3.88

6.00

8.97

1.50

2.26

3.17

4

Electrical equipment and computing

5.80

7.02

5.07

1.86

1.88

1.15

5

Motor vehicles

3.01

5.19

7.03

1.44

1.95

2.28

6

Rubber and plastic products

6.52

6.70

6.52

1.84

1.06

0.13

7

Non-metallic mineral products

2.70

6.64

12.85

1.13

1.56

2.24

8

Coal and petroleum products

2.03

2.12

2.26

0.52

0.51

0.75

9

Professional and scientific instruments

6.14

4.67

5.27

1.83

1.65

2.07

10

Other manufacturing

2.43

3.44

2.56

1.14

0.74

0.45

.

Total all industries

4.26

5.47

5.60

1.53

1.77

1.73


Source: As for Table 5.

 

As a result, the pattern of technological specialisation of US-owned affiliates in the UK has shifted away from electrical equipment and even to some extent from scientific instruments, and towards chemicals and pharmaceuticals, as shown in the RTA calculations for the more recent periods in Table 8. Indeed, at the very time that the UK's own technological specialisation has increasingly favoured R&D in chemicals and pharmaceuticals industry, from a RTA value of 1.11 in 1967-77 to one of 1.54 in 1987-95, the equivalent RTA of US-owned affiliates located in the UK jumped even more sharply from 0.98 to 1.73 (see Table 8). The UK having lost its revealed technological advantage in the electrical equipment industry by 1969 (which early post-war advantage in any case, as argued above, had had much to do with the remaining benefits of the early research-based investments of US TNCs in the UK in this industry during the interwar period), even US-owned electrical equipment affiliates in the UK gradually did comparatively less and less local research, falling from a RTA of 1.26 in 1969-77 to 0.62 in 1987-95. Thus, US-owned UK-located affiliates have moved away from the exploitation of the main areas of US home country strength (although the US electrical equipment TNCs have lost advantage even at home in recent times to the leading Japanese firms in this industry), and instead towards the industry in which indigenous UK firms are strongest - namely, in chemicals and pharmaceuticals. Beyond the latter case, one can also observe a growing US TNC focus in the UK on local development in the food product and non-electrical machinery industries (in which areas between 1969-77 and 1987-95 the US-owned affiliate RTA rises from 0.88 to 1.52, and from 0.96 to 1.58 respectively), the overall UK RTA in 1987-95 being 2.90 in food products and 1.07 in non-electrical machinery.

Table 8: Revealed technological advantages across industries of US and British firms, 1969-95.

. .

US affiliates located in the UK

.
.

UK as a host country

.
.

US firms in the US

.
.

.

.

1969-77

1978-86

1987-95

1969-77

1978-86

1987-95

1969-77

1978-86

1987-95

1

Food, drink and tobacco

0.88

1.26

1.52

1.59

2.68

2.90

1.06

1.09

1.29

2

Chemicals and allied

0.98

1.32

1.73

1.11

1.31

1.54

0.86

0.92

0.98

3

Non-electrical machinery

0.96

1.21

1.58

0.90

1.03

1.07

0.98

0.95

0.86

4

Electrical and computing

1.26

1.06

0.62

0.79

0.73

0.68

1.04

1.00

0.92

5

Motor vehicles

0.77

0.60

0.80

1.85

1.21

0.91

0.82

0.55

0.60

6

Rubber and plastic

1.30

0.71

0.08

1.97

1.44

0.75

1.08

1.19

1.14

7

Non-metallic materials

0.89

1.12

1.78

0.95

1.01

0.95

1.21

1.27

1.37

8

Coal and petroleum

0.40

0.42

0.64

0.70

0.66

1.11

1.19

1.46

1.47

9

Instruments

1.22

0.72

1.02

0.27

0.18

0.29

1.02

0.77

0.85

10

Other manufacturing

0.93

0.58

0.45

0.95

1.04

1.24

1.25

1.38

1.71


Source: As for Table 5.

 

Conclusions

US-owned research in the UK is comparatively concentrated in the chemical and pharmaceutical industry, and has reinforced the UK as a centre of excellence for R&D in that industry by contributing to the upgrading of the already existing innovatory, research and technological capabilities of indigenous British firms (Cantwell 1989, Cantwell and Hodson 1991, Cantwell, 1992). In some industries an increase in US-owned technological development has gone hand in hand with a rise of indigenous activity (in chemicals and pharmaceuticals, food products and to a lesser extent in oil), but in others (in particular in electrical equipment) there has been a process closer to a vicious cycle of mutual decline (Cantwell 1987). The growing correspondence between the industrial composition of the technological development of US-owned TNCs in the UK and the pattern of R&D in locally-owned companies is suggestive as well of an historical shift towards a strategy of international technology sourcing within US-owned TNCs, or what in his latest writings Dunning (1995) has stylised as asset-seeking investment. In turn, the rising significance of cross-border technology sourcing implies the establishment of a more complex and creative international corporate network for technology development within the TNC. Not surprisingly, this historical change has been a major theme in the later work of both Dunning and Vernon.

By the late 1970s Vernon (1979) recognised that the emergence of global structures within TNCs had eroded the explanatory power of the product cycle model. The model had been especially effective in depicting one particular historical phase of TNC development, namely the form of expansion of US TNCs in the early post-war period. It also proved useful as a means of emphasising the role of corporate technology leadership in international trade and investment, although leaders are now marked out not so much by their greater propensity (and capacity) to engage in foreign direct investment, but rather by their greater ability to organise cross-border networks for further technology development (Cantwell and Piscitello, 1999a). Yet when recast in this slightly different context of increasingly globally interconnected TNCs, another of Vernon's early themes in the product cycle model has reappeared as of great relevance today, and is now attracting a large literature. That is the theme of the locational agglomeration of innovative activity in the regional centres of TNCs, which of course Vernon had supposed to apply mainly to centres within the home country. Instead, now that the leading TNCs are operating across several geographically dispersed centres, technological development has become concentrated in certain key regions within countries, and TNCs source knowledge from each of them and provide an institutional means for the exchange of knowledge between the principal centres for development in each industry (Cantwell and Iammarino, 1999). Thus, much of Dunning's recent research has been concerned with the linkages between location-specific lines of technological innovation in regional centres and the global sourcing of technology and hence the creation of competitive advantage by TNCs (Dunning, 1996, 2000). Not only did Dunning and Vernon set the stage for the massive subsequent literature on technological innovation and the dynamism of the TNC, and provide many of the key insights that were to become crucial to later researchers, but they have both also proved adept at adapting their thinking to accommodate the major institutional changes that have occurred since their earlier investigations. Perhaps the most important such change has been the wider extent of technogically creative international networks within TNCs, which is well illustrated by the evolving structure of US-owned local technology development in the UK.

 

References

Acs, Zoltan J. and Audretsch, David B. (1989), ‘Patents as a measure of innovative activity’, Kyklos, Vol.42, pp. 171-180.

Cantwell, John A. (1987), ‘The reorganisation of European industries after integration: selected evidence on the role of multinational enterprise activities’, Journal of Common Market Studies, 26, pp.127-151.

Cantwell, John A. (1989), Technological Innovation and Multinational Corporations, Oxford: Basil Blackwell.

Cantwell, John A (1992), ‘The internationalisation of technological activity and its implication for competitiveness’, in Granstrand, O., Håkanson, L. and Sjölander, S. (eds.), Technology Management and International Business: Internationalization of R&D and Technology, John Wiley and Sons.

Cantwell, John A. (1995), ‘The globalisation of technology: what remains of the product cycle model?’, Cambridge Journal of Economics, 19, pp. 155-74.

Cantwell, John A. and Barrera, M. Pilar (1998), 'The localisation of corporate technological trajectoriesin the interwar cartels: cooperative learning versus an exchange of knowledge', Economics of Innovation and New Technology, 6, pp. 257-290.

Cantwell, John A., Dunning, John H. and Janne, Odile E.M. (1999), 'US direct investment in the UK: international systems of innovation revisited', mimeo, University of Reading.

Cantwell, John A and Hodson, Christian (1991), 'Global R&D and UK competitiveness', in Casson M.C. (ed), Global Research Strategy and International Competitiveness, Oxford: Basil Blackwell.

Cantwell, John A. and Iammarino, Simona (1999), 'Multinational corporations and the location of technological innovation in the UK regions', Regional Studies, 34, forthcoming.

Cantwell, John A. and Janne, Odile E.M. (1999), ‘Technological globalisation and innovative centres: the role of corporate technological leadership and locational hierarchy’, Research Policy, 28, pp. 119-144.

Cantwell, John A. and Piscitello, Lucia (1999a), 'The emergence of corporate international networks for the accumulation of dispersed technological competences', Management International Review, 39, Special Issue, pp. 123-147.

Cantwell, John A. and Piscitello, Lucia (1999b), 'Accumulating technological competence - its changing impact upon corporate diversification and internationalisation', Industrial and Corporate Change, 9, forthcoming.

Casson, Mark C. (ed, 1991), Global Research Strategy and International Competitiveness, Oxford: Basil Blackwell.

Chandler, Alfred D. (1990), Scale and Scope: the Dynamics of Industrial Capitalism, Cambridge, Mass.: Harvard University Press.

Dunning, John H. (1958), American Investment in British Manufacturing Industry, London: Allen and Unwin.

Dunning, John H. (1970), Studies in International Investment, London: Allen and Unwin.

Dunning, John H. (1994), 'Multinational enterprises and the globalisation of innovatory capacity', Research Policy, 23, pp. 67-88.

Dunning, John H. (1995), 'Reappraising the eclectic paradigm in an age of alliance capitalism', Journal of International Business Studies, 26, pp. 461-491.

Dunning, John H. (1996), ‘The geographical sources of competitiveness of firms. Some results of a new survey’, Transnational Corporations, 5, pp. 1-21.

Dunning, John H. (1998), American Investment in British Manufacturing Industry. Revised and Updated Edition, London: Routledge.

Dunning, John H. (ed, 2000), Regions, Globalization and the Knowledge Economy: The Issues Stated, forthcoming.

Dunning, John H. and Pearce, Robert D. (1985), The World's Largest Industrial Enterprises, 1962-1983, Farnborough: Gower.

Florida, Richard (1997), ‘The globalization of R&D: results of a survey of foreign affiliated R&D laboratories in the USA’, Research Policy, 26, pp. 85-103.

Fors, Gunnar (1998), ‘Locating R&D abroad: the role of adaptation and knowledge-seeking’, in Braunerhjelm, P. and Ekholm, K. (eds), The Geography of Multinational Firms, Dordrecht: Kluwer.

Granstrand, Ove, Håkanson, Lars and Sjölander, Sören (eds, 1992), Technology Management and International Business: Internationalization of R&D and Technology, Chichester: John Wiley and Sons.

Griliches, Zvi (1990), 'Patent statistics as economic indicators: a survey', Journal of Economic Literature, 18, pp. 1661-1707.

Kümmerle, Walter (1996), 'The drivers of foreign direct investment into research and development: an empirical investigation', Harvard Business School Working Paper, 96-062.

Lall, Sanjaya (1979), 'The international allocation of research activity by US multinationals', Oxford Bulletin of Economics and Statistics, 41, pp. 313-331.

Mansfield, Edwin, Teece, David J. and Romeo, Anthony (1979), 'Overseas research and development by US-based firms', Economica, 46, pp. 187-196.

Patel, Pari (1995), 'Localised production of technology for global markets', Cambridge Journal of Economics, 19, pp. 141-153.

Patel, Pari and Pavitt, Keith L.R. (1991), 'Europe's technological performance', in Freeman, C., Sharp, M. and Walker, W. (eds) Technology and the Future of Europe: Global Competition and the Environment in the 1990s, London: Frances Pinter.

Pavitt, Keith L.R. (1988), 'Uses and abuses of patent statistics', in van Raan, A. (ed), Handbook of Quantitative Studies of Science Policy, Amsterdam: North Holland.

Pearce, Robert D. (1997), ‘Decentralised R&D and strategic competitiveness: globalised approaches to generation and use of technology in MNEs’, Research Policy, 28, pp. 157-178.

Pearce, Robert D. and Singh, Satwinder (1992), Globalising Research and Development, London: MacMillan.

Soete. Luc L.G. (1987), ‘The impact of technological innovation on international trade patterns: the evidence reconsidered’, Research Policy, 16, pp.101-130.

Vaupel, J.W. and Curhan, J.P. (1973), The World's Multinational Enterprises: A Sourcebook of Tables, Boston, Mass.: Harvard Business School Division of Research.

Vernon, Raymond (1966), 'International investment and international trade in the product cycle', Quarterly Journal of Economics, 80, pp. 190-207.

Vernon, Raymond (1979), 'The product cycle hypothesis in a new international environment', Oxford Bulletin of Economics and Statistics, 41, pp. 255-267.

Wilkins, Mira (1974), The Maturing of the Multinational Enterprise: American Business Abroad from 1914 to 1970, Cambridge, Mass.: Harvard University Press.