This article was downloaded by:[Middle East Technical University]

[Middle East Technical University]

On: 25 February 2007
Access Details: [subscription number 738903067]
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Oxford Development Studies

Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713439972
Innovation in Asian Industrialization: A Gerschenkronian
Perspective
Mike Hobday

To link to this article: DOI: 10.1080/1360081032000111715

URL: http://dx.doi.org/10.1080/1360081032000111715

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction,
re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly
forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents will be
complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be
independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or
arising out of the use of this material.
© Taylor and Francis 2007
 Oxford Development Studies, Vol. 31, No. 3, September 2003

Innovation in Asian Industrialization: A

Gerschenkronian Perspective

MIKE HOBDAY*
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

ABSTRACT This paper interprets the experience of the East and South East Asian electronics
industry from a “Gerschenkronian” perspective in order to draw lessons for other developing
countries. Following Gerschenkron, the paper argues that it is diversity, rather than unifor-
mity, in the institutional, technological and development policy arenas (called here “strategic
innovation”) that characterizes the experience of the Asian newly industrializing economies
(NIEs). The experience of the leading export industry shows that the progress of the NIEs can
be interpreted as a pattern of substitution of missing prerequisites, in line with Gerschenkron’s
view of European latecomer industrialization. More broadly, the progress of the NIEs should
not be viewed as repetitions of earlier industrialization experiences as they involve significant
deviations from the latter, usually entailing distinctive strategic innovations. This interpret-
ation presents a difficult challenge for those wishing to draw lessons from the Asian NIEs.
There are few direct lessons from East and South East Asia for other countries and certainly
no transferable or standardized “model” of development. Other paths and patterns of develop-
ment need to be identified and created that build upon the distinctive resources of individual
developing countries. Strategic innovation, trial-and-error learning and variety are likely to
continue to characterize successful industrial development in the future.

1. Introduction
Innovation, at the institutional, technological and development strategy levels, is an
important issue in the context of economic development.1 Indeed, based on research
into the economic development of Europe, the historian Gerschenkron argued that
innovation is at the heart of economic progress at the national, industrial and wider
institutional levels. However, this proposition has yet to be assessed in the context of
recent East and South East Asian economic development
The aim of this paper is to interpret East and South East Asian industrialization in
the light of Gerschenkron’s (1962, p. 359) argument “that European history should be
seen as a pattern of substitution governed by the prevailing—and changing—degree of
backwardness”. By examining the leading export sector, electronics, the paper assesses
the extent to which the development of Asia’s newly industrializing economies (NIEs)
can be viewed as a pattern of strategic innovation, involving the substitution of earlier
development patterns and factors with novel paths and alternative “prerequisites” for
industrialization.

* Mike Hobday, SPRU: Science Technology and Policy Research, University of Sussex, Falmer, Brighton
BN1 9RF, UK.

ISSN 1360-0818 print/ISSN 1469-9966 online/03/030293-22  2003 International Development Centre, Oxford
DOI: 10.1080/1360081032000111715
 294 M. Hobday

For many years economists and technology researchers of various kinds have
analysed how developing countries might grow and even catch up by proceeding
through certain stages of development (e.g. Kuznets, 1954, 1966; Chenery, 1968;
Rostow, 1960; Gerschenkron, 1962, 1963; Kim, 1997).2 This paper assesses the extent
to which the progress of the Asian NIEs should be seen as a series of repetitions of
earlier industrialization stages, or as significant deviations from earlier experiences,
entailing distinctive strategic and institutional innovations in individual countries. This
is of interest to the general understanding of the industrial development of the Asian
NIEs and for exploring the role of innovation in that development. It may also be useful
for other developing nations wishing to draw lessons from their experiences.3
To address these issues, Section 2 briefly summarizes Gerschenkron’s view of
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

latecomer development focusing on the need for the substitution of missing prerequi-
sites by late industrializers. Section 3 examines the growth of the electronics industry
in East Asia, interpreting the historical stages witnessed by researchers from a Ger-
schenkronian perspective.4 Section 4 turns to the South East Asian NIEs and contrasts
their development paths with those of the East Asian NIEs. Section 5 reflects on these
experiences in order to identify the mechanisms and degree of substitution of missing
prerequisites in the Asian NIEs. The conclusion then draws lessons for other develop-
ing countries, taking into account Gerschenkron’s warnings in this area.

2. A Gerschenkron Perspective on Latecomer Development

In his critique of Rostow (1960), Gerschenkron argued that the study of industrial
revolutions of the past could not provide a sensible guide for today’s policy-makers
wishing to promote economic development. In particular, he emphasized that the idea
of identifying which prerequisites (or preconditions) were “missing” in a particular
developing economy (e.g. in terms of investment, institutions and technology) and then
installing them was not a logical strategy for achieving a take-off to sustained growth.
Although Gerschenkron was interested in explaining what he called “the great
spurts” of industrialization and viewed the course of industrialization as a schematic
stage-like process, he viewed the argument that industrialization did or should proceed
through certain more or less discrete stages with a great deal of suspicion and
scepticism.5 He argued that the idea that every development case should be based on
the same set of preconditions was a classic example of historical determinism (Ger-
schenkron, 1962, pp. 356–358).
Gerschenkron argued that there were no automatic stages of development and that
countries did not and could not pass through the same stages of development that others
had passed through before them, precisely because others had passed through them. As
a result, barriers to entry and opportunities for development would have changed. Each
latecomer economy would therefore face a very different external environment in terms
of markets, technologies and opportunities for growth.
Referring to his own interpretation of historical stages, he argued that:

this [interpretation] differs essentially from the various efforts in “stage mak-
ing”, the common feature of which was the assumption that all economies
were supposed regularly to pass through the same individual stages as they
moved along the road of economic progress … Thus, Rostow was at pains to
assert that the process of industrialization repeated itself from country to
country lumbering through his pentametric rhythm.
(Gerschenkron, 1962, p. 355)
 Innovation in Asian Industrialization 295
In contrast to Rostow, and much conventional wisdom of the day, Gerschenkron
argued that it is essential to look at the differences in stages between individual
countries and to analyse in detail the dynamic processes of change. He focused on the
role of different institutions (e.g. the function of banks in central European industrial-
ization and the importance of the Ministry of Finance in Russian industrialization) in
promoting industrialization. Regarding general prerequisites for industrial growth (e.g.
the provision of social overhead capital and a favourable “value system”), through his
own research on latecomer development, Gerschenkron showed that these prerequisites
were sometimes not present or only present to a limited extent in latecomer economies
which had gone on to develop. Indeed, in some cases the great “spurt” occurred
without them.
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

Gerschenkron’s view was that latecomers had to plot their own distinctive path of
development. These paths had to take account of how other earlier developers had
progressed, including the markets and technologies they had created and influenced.
He also argued that the starting position of the latecomer economy (or stage of
backwardness as Gerschenkron called it) had to be taken into account. In particular, his
research showed how the particular conditions of backwardness shaped the nature,
depth and path of state intervention with respect to enterprise, technology, institution
building, resource mobilization and so on.
The views of Gerschenkron have important implications for innovation and devel-
opment. They imply that development is an active “bottom up” process involving
economic action and reaction, as well as heavy investments in learning, technology and
the institutions of development. As many observers have emphasized, this type of
purposeful strategy contrasts vividly with market-led processes or passive “trickle
down” approaches to development.6 Hoping simply to “pass through” the stages of
development that earlier developers had followed was highly unlikely to bring about
development. Indeed, Gerschenkron emphasized the importance of variety and differ-
ence in the development paths of nations.
In Gerschekron’s model, strategic innovation (i.e. doing things new to the world or
the market-place) in relation to government policy, development path, technology
acquisition, institution building and so on, is a central part of economic development.
Only by choosing and successfully following distinctive paths (and therefore stages) of
development can latecomer nations meet the new circumstances presented to them by
the actions of earlier developers. While he also recognized that any development path
followed will not be entirely new, it will have to embody at least some innovative
features to cope with the new environment. In this way he linked innovation to the
development process in general.
The next two sections of the paper turn to the recent research literature on the
stages of technological and industrial development of the economies of East and South
East Asia.7 These success cases appear to confirm, with some reservations, the import-
ance of innovation not just in the technological sense but also in the wider industrial
and institutional sense of government policy, paths of development, institutional
creation and novel firm-level strategies.

3. Stages of Industrial and Technological Development in East Asia

There is today abundant evidence on the performance of the NIEs of East and South
East Asia over the past 30 years or so. However, while there are major ongoing debates
over how to interpret the evidence (e.g. for various views on the “state versus market”
 296 M. Hobday

debate, see Amsden, 1989; Wade, 1990; World Bank, 1993; Hong, 1997), there is
relatively little discussion of how to interpret the technological and industrial stages
passed through by these countries or whether or not one can draw lessons from these
stages. This section therefore presents some of the key development achievements of
the region and then goes on to assess East Asian stages of development, focusing on
electronics.

3.1 “Spurts of Industrialization” in the Asian Region

The growth performance of the NIEs (Korea, Taiwan, Singapore and Hong Kong)
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

over the past 30 years or so is remarkable. It is worth remembering that as recently as

1962 Korean GDP per capita levels were on a par with the poorer African nations such
as Zaire and the Congo. Poverty was endemic, there was little industry in the South and
very high rates of illiteracy. Taiwan fared little better. Both Korea and Taiwan were
largely agricultural economies. However, over the next three decades both economies
grew at unprecedented rates and, by 1986, they had progressed up the rankings of
nation by 47 and 55 places, respectively, having registered GDP growth rates of
between 8 and 10% per annum,8 overtaking other developing countries such as Mexico
and Brazil (Wade, 1990, pp. 34–35).
The two smaller economies, Hong Kong and Singapore, also suffered from poverty,
unemployment and poor housing in the early 1960s. Both had little in the way of
manufactured exports and lacked industrial capabilities. Although Hong Kong and
Singapore began at relatively high per capita levels, they too forged ahead by 12 places
and 13 places, respectively, overtaking Brazil, Mexico, El Salvador and South Africa,
among others.
As a result of rapid growth, until the financial crisis of 1997 each of the four NIEs
approached full employment, with rising wages and increasing technological intensity in
production and exports. In each case, low-cost assembly was progressively moved out
of the home economy to lower wage locations in China, Malaysia, Indonesia, Thailand
and Vietnam, leading to rapid growth in these countries. Over a 35-year period a
significant proportion of world manufacturing and trade shifted from the USA and
Europe to the East and South East Asian region, partly as a result of the upgrading of
manufacturing skills and capabilities. In addition, in each of these countries (except for
Hong Kong, where clothing remained the largest export industry), electronics grew to
become by far the largest export sector.
We can reasonably interpret these achievements as the “great spurts” to which
Gerschenkron referred. These growth rates remain unprecedented in the post-war
developing world and it is not surprising that many scholars and policy analysts have
attempted to draw lessons for other countries and other regions (e.g. World Bank,
1993; Wade, 1990; Amsden, 1989; Kim, 1997).
Within this overall growth, the electronics industry was one of the leading modern
industrial sectors, representing a powerful engine of economic transformation. As a lead
sector, it introduced new production functions of high productivity, raising output per
capita. It diverted investment to firms which reinvested profits in productive investment
and established a chain of demand for other manufactured products. Along with other
modern sectors it required capital upgrading and new organizations to promote exports.
The electronics industry played an important role in enabling the East and South East
Asian economies to connect to export markets and accelerated the inward transfer of
foreign technology.
 Innovation in Asian Industrialization 297
3.2 East Asian Technology Paths: South Korea, Taiwan and Hong Kong
At the start of their rapid growth period, South Korea, Taiwan and Hong Kong faced
severe latecomer disadvantages, especially in gaining access to foreign markets and
technology.9 In particular, they were largely disconnected from the main international
sources of technology, which tended to be located within the most developed countries
(e.g. various parts of the USA, Japan and Europe). In addition, the NIEs were for the
most part disconnected from the developed country markets they hoped to export to,
having few and weak linkages with marketing and distribution channels, especially in
the USA and Europe.
If we examine the growth in electronics exports, by far the most important institu-
tional mechanism that evolved for exporting from East Asia was the so-called original
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

equipment manufacture (OEM) system.10 This was extremely important for South
Korea, Taiwan and Hong Kong in their efforts to export to international markets and
to acquire foreign technology.11 OEM is a specific form of subcontracting under which
a complete, finished product is made to the exact specification of the buyer (often a
large transnational corporation (TNC)). The product is then marketed through the
buyer’s own distribution channels and brand name. Typically, the TNC will transfer
the necessary technology to the latecomer firm in order to gain the advantages of
low-cost labour. The TNC gains all the non-manufacturing value added (e.g. design,
R&D and distribution), which typically exceeds the value added in production, thus
earning large returns and creating competitive advantages for foreign firms. Once one
major TNC engaged in OEM (e.g. Texas Instruments in semiconductors in the early
1960s), other TNCs quickly followed to gain similar advantages and avoid falling
behind.

3.3 A Simple “Stages” Interpretation

If we turn to empirical research into exporting firms over the past 30 years or so, in line
with Gerschenkron’s view, the use of stages as a historical device in some ways is quite
helpful in summarizing how the OEM system operated, evolved and expanded through
time, functioning both as an institutional mechanism for acquiring technology and for
accessing export marketing channels (see Table 1).
Research reveals several key points about the OEM system. First, it began experi-
mentally in fairly small ventures in the 1950s and early 1960s, pioneered by US firms
including semiconductor producers (e.g. Texas Instruments and IBM). Under the
OEM system, latecomer subcontractors were able to gain economies of scale and learn
basic production operating techniques. Korean firms such as Samsung and Goldstar
(later Luck-Goldstar, then LG) gained new routes into US and, later, European export
markets. Factory workers, technicians, engineers and managers were trained in modern
technology, management and sometimes accountancy techniques, by the large TNC
buyers who wished to gain the advantages of low-cost labour, land and other operating
costs.
In the 1980s, a new variant of OEM called own design manufacture (ODM)
evolved, first noted by Johnstone (1989) for the case of Taiwan. Under ODM, the
developing country firm carries out most of the detailed product design, having received
an overall design layout from the TNC customer. The TNC continues to carry out the
marketing and distribution functions under its own brand name, thereby continuing to
gain most of the value added. However, the latecomer firm is expected to develop
 298 M. Hobday

Table 1. Stages of latecomer development: from OEM to ODM

to OBM

Technological transition Market transition

1960s/1970s Local firm learns Foreign TNC/buyer

OEM assembly process for designs, brands and
Original equipment standard, simple goods distributes/gains non-
manufacture manufacturing value
added

1980s Local firm learns As with OEM, TNC

ODM process engineering buys, brands and
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

Own design and and detailed product distributes. TNC gains

manufacture design skills non-manufacturing
value added

1990s Local firm conducts Local firm has own

OBM manufacturing, product brand, organizes
Own brand designs and conducts distribution and
manufacture R&D for new products captures all value
added

Source: Amended from Hobday (1994).

minor product improvement skills and to set up the necessary, often large-scale,
manufacturing processes.
Then, in the 1990s, some of the leading firms in East Asia began own brand
manufacture (OBM), competing directly with international suppliers from Japan, the
USA and sometimes Europe (e.g. Philips of Holland in consumer goods). Under OBM,
the latecomer firm carries out all of the stages of production and innovation, including
manufacturing, new product designs and R&D for materials, processes and products.
At this advanced stage there is little difference between a latecomer and a leader (or a
follower). The latecomer firm has typically developed its own brand, organizes its own
distribution abroad, carries out R&D and is able to capture all of the value added
associated with production, branding and distribution.

3.4 A Gerschenkronian Interpretation of Stages

How should we interpret these apparent stages and what are their strengths and
weaknesses as an analytical device? Regarding strengths, the stages approach is quite a
useful method for describing broad historical trends (including technology and market-
ing processes) in the industrial development of the electronics industries. Indeed,
Gerschenkron used stages in this way in his writings although, of course, he was mainly
talking about much greater sweeps of industrial history and entire economies rather
than leading sectors.
Stages, and the research underpinning them, appear useful for summarizing some of
the key features of technological innovation in the context of a catching-up firm or
economy. What the stages indicate is the progressive move from simple tasks to more
complex, more technology-intensive activities, over long periods of time. Indeed, case
studies of firms engaged in OEM show that many firms gained technology in a gradual
step-by-step incremental manner over periods of 20 or even 30 years (Cyhn, 2002;
Hobday, 1995).
 Innovation in Asian Industrialization 299
The stages model also indicates that the core technological activity involved for
many years in the catch-up process is not R&D-centred innovation (e.g. for new
products or processes) or radical innovation of any kind. By contrast, radical innovation
is a major concern of the western firms such as Microsoft and IBM and leading
Japanese firms.12 Latecomers, instead, need to “keep up” or “catch up” through small,
incremental improvements to existing products and processes and this requires engin-
eering and technician skills rather than R&D. What the research reveals is the import-
ance of minor improvements to products, the changing of processes to become more
efficient and more flexible and, later on, the introduction of new variations of product.
The nature of technological innovation in these latecomer firms provides a new
approach to technology acquisition, very much in line with Gerschenkron’s view of the
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

need to introduce new patterns of change. The traditional definition of technological

innovation, namely producing a new (or improved) product in the market-place or to
the world, usually based on R&D, in East Asia is replaced by a range of “behind the
frontier” incremental innovations, including improvements to products, the changing
of processes to become more efficient and flexible, and the introduction of new types
of product based on the designs of leading firms.
Furthermore, the stages research captures the fact that innovation occurs, not just
in technological terms but also, and again in line with Gerschenkron, in institutional
terms. The technological innovation that took place in East Asia in electronics probably
could not have occurred with such rapidity without the preceding and accompanying
organizational innovations of OEM, and later ODM and OBM. Basic OEM was
originally a marketing term that referred to the subcontracting arrangement between
the supplier and the buyer. However, as shown above, the OEM system also had a very
important, if latent, technological function. It represented the institutional mechanism
that enables exports and technology transfer to take place in tandem, and it eventually
grew to become a very important institutional form within East Asia.
To sum up, as an analytical method for organizing 30 years of historical develop-
ment in one leading sector, electronics, the stages model is quite useful. However, used
in isolation or incorrectly, the stages approach has various disadvantages, alluded to by
Gerschenkron. As a result, if it is not used correctly and in combination with other
development information it can actually conceal more than it reveals.
Turning to weaknesses and limitations, first, as with any historical generalization,
one cannot push the scheme too far. As with any broad description, it is a gross
simplification of reality. The processes outlined are certainly not true for every firm,
even within electronics, or necessarily correct for every electronics sub-sector. For
example, telecommunications exchanges and higher cost computers did not follow this
path, nor did industrial electronics (e.g. numerically controlled machine tools in
Korea). The stages model applies only to low-cost, relatively simple telecommunica-
tions and computer products, not to “high end” goods. The stages model also masks
significant differences between exporters. For example, major Korean firms such as
Samsung began investing in R&D very early on and, in some product areas (e.g.
microwave ovens), as early as the 1960s, long before they progressed to OBM.
Other firms (e.g. Acer of Taiwan) “jumped in” at later stages, benefiting from the
technological infrastructure, institutions and other advances made possible by the early
entrants. It is by no means true to say that all East Asian firms entered in the 1960s and
steadily marched forwards as the stages approach might suggest. Indeed, many firms
(e.g. in Taiwan) failed and went bankrupt or, in some cases, developed financial
problems and were taken over by other firms.
Furthermore, even today at a fairly advanced stage of catch up, many of these East
 300 M. Hobday

Asian firms conduct a mixture of OEM, ODM and OBM. It is not correct to imply, as
the model might, that all firms have progressed to OBM. Indeed, even the most
advanced producers such as Samsung of South Korea still produce large quantities of
products under basic OEM arrangements (e.g. in consumer electronics) with Japanese
and, in some cases, US firms. Leading Korean and Taiwanese firms have gained
valuable experience in how to locate basic low-cost OEM activities in low-cost locations
within China and other countries (including Eastern Europe and Mexico). At the same
time, they now compete at the technology frontier with Japanese, US and European
market leaders (e.g. with third-generation mobile phones and DRAM semiconductors)
backed up by very large R&D investments. Today, firms such as Samsung of Korea and
ACER of Taiwan could more accurately be described as “hybrid” firms, containing
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

features of leadership in some product areas (e.g. advanced R&D), “followership” in

other areas (i.e. having R&D capability but following others quickly into markets) and
basic latecomer activities in still other areas.
So, in line with Gerschenkron’s view, even as a historical device for analysing a
single sector, in this case electronics, the stages model is in some respects a misleading
simplification. Furthermore, if one looks beyond electronics, detailed research shows
that other products and sectors do not follow the pattern (e.g. in capital goods
including aircraft and nuclear power equipment).13 In these areas, subcontracting does
sometimes occur but it has taken a very different path. In aircraft engines, nuclear
power equipment and capital goods more generally (e.g. telecommunications ex-
changes), Korean producers have been far less successful in developing technology and
matching the capabilities of international suppliers. Even in personal computers
Korea’s chaebol have had little success until recently in international markets, despite
major efforts (Hwang, 1998).

3.5 Innovation in Policy and Industrial Structures

Following the search for variety and innovation emphasized by Gerschenkron, it is clear
that, despite the apparent similarities in stages of development of firms and industries,
the preconditions for growth (including government policies and industrial structures)
were dissimilar among the different NIEs.
Figure 1 illustrates the variety of policy models and industrial structures for
electronics in the three East Asian economies, with Singapore added for further
contrast. The vertical axis refers to the degree of direct policy intervention, while the
horizontal axis refers to the degree of openness of each economy. As Figure 1 shows,
the vertical axis indicates a high degree of direct government intervention in the early
stages of development of electronics (i.e. in the 1960s and 1970s) for South Korea and
Singapore but shows a lower degree of direct intervention in Taiwan and a more or less
“laissez-faire” approach in Hong Kong. In South Korea, for example, the Government
directly supported the large chaebol with favourable finance, subsidies and many other
privileges (Amsden, 1989). Indeed, some authors suggest the Government helped
create the chaebol as we now know them (Jones & Sakong, 1980)
In Singapore, the Government also supported large firms, but in this case foreign
companies. The Government, led by Lee Kuan-yew, believed that the local en-
trepreneurial base was too weak to lead industrialization and intervened to attract large
manufacturing TNCs, offering them incentives, infrastructure in the form of export
processing zones, and a degree of freedom rarely seen before in the developing world
(Yue, 1985). By contrast, much of Taiwanese early development in electronics was
based on small and medium-sized companies, linked to international markets through
 Innovation in Asian Industrialization 301
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

Figure 1. Diversity of policy models: the four NIEs. Source: Hobday (1995, p. 196).

the multitude of traders in that country (Chou, 1992). Many of these firms were fearful
of government and some were more or less in the “underground” economy.14 Only later
on, and only in certain areas, did the Taiwanese Government become more directly
involved in supporting electronics producers (e.g. in Hsinchu and other industrial parks
and with institutions such as the Information Technology Research Institute in semi-
conductors; Matthews & Cho, 2000).
As the horizontal access of Figure 1 indicates, Korea and Taiwan were fairly closed
to foreign direct investment (FDI) for much of the 1960s, 1970s and 1980s. By
contrast, there was far greater FDI in Singapore and Hong Kong despite their smaller
GDPs. For example, taking the two largest investors, Japan and the USA, during the
1980–88 period total FDI amounted to US$14.3 billion in the four NIEs. Hong Kong
received the largest amount of FDI (US$6.3 billion); Singapore came second with
US$3.6 billion; South Korea received only US$2.3 billion; and Taiwan US$2.1 billion
(James, 1990, p. 15). Singapore and Hong Kong placed few restrictions on FDI.
Conversely, Taiwan and South Korea tightly controlled FDI and protected local
industries from foreign competition, encouraging domestic firms to supplant foreign
ones wherever possible.15
Figure 1 also points to variety in the orientation of industrial policy. While Hong
Kong and Singapore pursued conventional export-led policies, South Korea and
Taiwan combined these policies with import-substitution, controlling or banning im-
ports to protect local firms and using government procurement to stimulate local
 302 M. Hobday

enterprise. South Korea was very restrictive, banning most consumer goods and raw
materials that did not have to be imported. The Taiwanese Government often negoti-
ated the terms of FDI and tied TNCs to local content rules and export targets.
Figure 1 notes important differences in company size and ownership which were
both shaped by and flowed from government policies. While Taiwan and Hong Kong
depended to a large extent on small (and a few large) Chinese-owned family businesses,
the South Korean Government patronized the large conglomerates. South Korean
policies resulted in highly concentrated industrial structures, with the chaebol dominat-
ing electronics and many other industries. By contrast, in Hong Kong and Taiwan
small firms proliferated and grew, resulting in a highly dispersed industrial structure in
electronics.16 When comparing industrial concentration, it is difficult to imagine a more
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

stark contrast than between South Korea and Taiwan

Policy and company strategy were also closely interconnected. In Taiwan their small
size led local firms to rely on speed and flexibility, while the large South Korean
companies gained advantages in high volume, process-intensive electronics sectors. By
contrast, many Taiwanese and Hong Kong firms specialized in fast changing market
niches, whereas South Korean corporate strategies in some respects imitated the
Japanese keiretsu, emphasizing scale and vertical integration. Regarding ownership,
Taiwan and South Korea relied mostly on locally-owned firms, while Singapore
depended almost entirely on foreign TNCs. Again in contrast, Hong Kong relied on a
mixture of local and foreign firms to lead electronics.
In short, by adopting a Gerschenkronian perspective, one can see striking contrasts
between the four NIEs in their industrial development, alongside strong similarities in
stages of development. Policy diversity was associated with major differences in indus-
trial concentration, corporate ownership and strategy, as well as patterns of innovation
and paths of industrial development. If we turn to the South East Asian economies (e.g.
Singapore, Malaysia, Thailand, Indonesia and the Philippines), we find a very different
development model in electronics, based on FDI and relying mostly on TNC sub-
sidiaries’ exports.

4. South East Asian Electronics Development

4.1 Performance of South East Asia
Before comparing the two different models, it is useful to look briefly at the export
performance of four of the leading NIEs. Table 2 presents information on GDP and
electronics exports for four countries (two East Asian, Korea and Taiwan, and two
South East Asian, Singapore and Malaysia) for a typical year in the mid-1990s. The
first point to note is that rather than South Korea (normally assumed to be the leading
NIE), it is actually Singapore (with a tiny population and much smaller GDP) that
leads in terms of electronics exports (US$38 billion in Singapore versus US$25 billion
for South Korea). Note also that, in Singapore, the value added in electronics makes a
sizeable contribution to overall value added, indicating that electronics exports are not
merely transhipments.17
Another surprise is the Malaysian performance. Exports are almost at the level of
South Korea, despite a much smaller GDP and a later start in electronics (Malaysia
began exporting during the 1970s, around 10 years later than Korea). In addition,
despite the greater attention usually given to the “Korean model”, Taiwan is not far
behind South Korea, despite its far greater dependence on small and medium-sized
enterprises (SMEs) and much smaller GDP.
 Innovation in Asian Industrialization 303
Table 2. Exports of electronics Korea and Taiwan versus Singapore and Malaysia
(mid-1990s)

Population Total GDP Total exports Electronics Share of

millions (1994 (US$ billions exports 1994 electronics in
(1995) US$ billions) 1994) (US$ billions) exports (%)

South Korea 45.6 508 96 25 26

Taiwan 21.5 257 93 18 19
Malaysia 19.7 167 57 22 39
Singapore 2.9 57 96 38 40
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

Overall, the strong performance of the two leading South East Asian countries
suggests that the TNC-led path may present a rival model of development in the overall
region. However, an immediate policy concern often raised in South East Asia is
whether or not (or to what extent) the TNCs actually transfer technology. Often, the
assumption is that the TNCs tend to be assembly operations (or “screwdriver” plants),
inferior to the locally-owned OEM path of development of South Korea and Taiwan.
To address the technological development issue in South East Asia it again appears
useful to use a simple stages model to summarize research on the patterns and
processes of technology development within the TNCs of South East Asia.

4.2 A Simple Stages Model of South East Asian Development in Electronics

Table 3 presents a simple stages model of South East Asian development in electronics
based on various studies of electronics in the region. Beginning with the earliest
developer, Singapore, under FDI the TNC subsidiaries gradually learned technology,
beginning with assembly operations in the 1960s, progressing to process engineering in
the 1970s and minor product improvements in the 1980s (Hobday, 1995). Recent
research shows an increase in R&D during the 1990s as wages rose and technical
competence improved (Wong, 1992, 1998). Often, R&D activities are associated with
the upgrading of TNC plants to regional headquarter status, with more companies
viewing Singapore as a business and technical support hub for the region. As in the case
of East Asian OEM, exporting within the TNCs enabled Singapore to overcome
technological and market barriers to entry in electronics, providing engineering training
for local staff, technology transfer, access to export markets and increasing economies
of scale.
Turning to the other countries of South East Asia, as Table 3 shows, the TNCs in

Table 3. Technological progress in South East Asia

Singapore Malaysia Thailand Indonesia Vietnam

1960s Assembly
1970s Process Assembly Assembly
engineering
1980s Product Process Assembly Assembly Assembly
development engineering
1990s R&D Product Process Process Assembly
development engineering engineering

Source: Based on empirical research (see text for references).

 304 M. Hobday

Malaysia began later (in the 1970s) with assembly. While there was nothing particularly
new in TNC technology transfer to these countries per se (e.g. to serve domestic
markets or to engage in tariff hopping), what was new was technology transfer to enable
exports. Technology transfer allowed for the rapid start up of exporting plants, the
expansion of existing investments, the attainment of rising levels of efficiency in
operations and the progressive upgrading of the type of products being exported from
each country.
In support of export-led growth, TNC subsidiaries engaged in technological learn-
ing at the plant level in order to acquire successively higher levels of technology.
However, Malaysia and the other South East Asian economies remained somewhat
behind Singapore through the 1980s and 1990s, lacking in R&D and new product
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

development capabilities (Ariffin & Bell, 1998; Bell et al., 1996; Rasiah, 1994). Within
the TNCs, Malaysia, and later Thailand, acquired manufacturing process skills and
some limited product design capabilities, beginning with simple tasks and gradually
moving towards more complex activities such as process adaptation and, in some cases,
limited R&D activities.18 Research also shows that Indonesia and Vietnam have experi-
enced export growth, but have yet to achieve the levels of capability development
witnessed by the earlier South East Asian countries.19
Consistent with Gerschenkron’s view of historical stages, the above interpretation
appears to be quite useful for summarizing various features of innovation in South East
Asia. For example, regarding technology, the TNC subsidiaries invested in technical
and engineering skills to assimilate and improve on existing technology, rather than to
move the innovation frontier forward through R&D, in a manner very similar to trends
within the OEM system of East Asia. In-depth firm-level studies (e.g. in Malaysia)
show that the subsidiaries had to struggle over many years to acquire technology and
that the management skills and performance of local plants were important consider-
ations in technology transfer.20 This, in turn, implies that the notion of “technology
transfer” (from parent to subsidiary) can be misleading, as the key factor is local
capability building on the part of the subsidiary, without which technology transfer
cannot occur. The term “technology acquisition” more accurately describes the process
at work.
Looking for the substitution of missing prerequisites that Gerschenkron empha-
sized, we can interpret the expansion of TNC-led growth and technology transfer in
support of exports as a significant institutional innovation. TNCs represented a new
organizational approach, allowing for the transfer of technology, systematic local
technological learning and the export of products to international markets. This
institutional innovation began in Singapore and was imitated by other countries wishing
to export to the USA and Europe. Although FDI had occurred prior to this, the
electronics industry represented a huge expansion of FDI in the South East Asian
region in direct support of exports (rather than for serving the local market). FDI
eventually led to the development of “clusters” of industrial development in various
locations outside the home country of the TNC headquarters. For example, today the
disk drive cluster in Thailand is the largest in the world. Similarly, in Penang, Malaysia,
the semiconductor packaging and testing industry has grown to become the largest
international concentration of this activity.
The emergence of TNC subsidiary exporting from small beginnings to large export
zones of production, strongly supported by governments in South East Asia, probably
ranks as important an institutional innovation as the OEM/ODM system described
earlier. Without the expansion of TNC investments the export of electronics and the
assimilation of technology by these countries probably would not have occurred
 Innovation in Asian Industrialization 305
although, of course, there is no way of knowing how alternative development paths
might have progressed.
Some observers have criticized the heavy reliance on TNCs as diminishing the
prospects for local firms, asking whether Singapore and Malaysia could have achieved
an even more impressive export performance if they had managed successfully to
promote local exporters, as in the case of Korea and Taiwan. One could also speculate
on whether, in East Asia, Korea would have grown even faster in electronics had it
allowed more FDI and especially, in more recent years, from the capital goods
producers for electronics.21
In South East Asia, as in East Asia, success was neither automatic nor “painless”.
Studies demonstrate long periods of trial-and-error experimentation, extensive training
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

of local operators, technicians and engineers and a great deal of human skill and
management ingenuity in the more successful cases of technology development within
TNC subsidiaries.
To sum up, TNC subsidiaries played a central role in South East Asian industrial
development in the electronics sector by providing a route into international markets
and by enabling continuous, routinized technological learning within local plants. The
“master–pupil” relationship described in studies of East Asian OEM (Cyhn, 2002)
mirrors the relationships that developed between parent and subsidiary plants in South
East Asia.

4.3 The Appropriate Use of Historical Stages

The South East Asian case shows that it is possible to summarize key historical features
of development using a stages interpretation. Stages, for some purposes, are useful for
analysing broad patterns of historical development and capturing some of the main
processes of technological and institutional innovation. However, the stages approach
suffers from many of the same problems as discussed in the case of OEM. First of all,
there are many exceptions at the level of firm and sub-sector, even within electronics.
Many firms did not progress in the way proposed. Indeed, individual TNC corporate
strategies towards centralization and decentralization had, and continues to have, a
major impact on the speed and direction of technical change in the region. Some
authors argue that Japanese firms, for example, are less willing to transfer technology to
their subsidiaries, while US firms take a more decentralized approach, giving high
performing subsidiaries more and more freedom to take their own technology deci-
sions.22
Second, as with OEM, ODM and OBM, each category in Table 3 exists side by side
at later stages of development. Even in the most advanced country, Singapore, as-
sembly and process engineering remain an important activity and R&D is only a
fraction of the technological activity carried out. Most R&D, of course, continues to
take place in the parent headquarters of the leading TNCs rather than in Singapore or
Malaysia. This applies especially to advanced materials, new product concepts, radical
new process designs and so on.
Third, it is certainly not automatic, as the table might imply, that follower countries
such as Thailand and Vietnam will make the transition to the more advanced stages.
They will no doubt experience many problems and setbacks as a result of the difficulties
of coping with more complex stages of technological development. In addition, as
Gerschenkron pointed out (see Section 2), circumstances facing latecomers constantly
change. For example, competition from major new entrants such as China and Mexico
could well upset or slow down the progress of later South East Asian industrializers,
 306 M. Hobday

unless they are able to offer competing benefits to TNCs choosing to locate abroad.
Indeed, China especially could well “upset” the neat stages model. As a major
latecomer with huge technical resources and an attractive future market potential to the
TNCs, China might well be able to catch up quickly and challenge the progress of
countries such as Thailand, Vietnam and Indonesia. In fact, research shows that China
is combining various institutional innovations (e.g. OEM, ODM, OBM, FDI and joint
ventures) to forge ahead and is using its potentially large local market to attract more
and more TNCs to the country.
In short, the case of South East Asia, like East Asia, illustrates both the uses and
potential abuses of stages models. In South East Asia there are many exceptions to the
rule of progress and usually each stage exists side by side in the countries examined. For
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

example, in Thailand in some electronics sectors many subsidiaries still conform to the
“screwdriver” plant view of assembly plant operation while in others (e.g. disk drives)
substantial technological progress has taken place centred on manufacturing inno-
vation. There is evidence that later entrants “jump in” and catch up much more quickly
(e.g. China) with new strategies. There is also considerable sub-sector specialization
within electronics (e.g. semiconductors in Malaysia and disk drives in Thailand).
Furthermore, there are significant differences in the policy approaches followed within
each country, with Singapore emphasizing direct government support to TNCs and
joint technology programmes, and Thailand, for example, offering far less in the way of
direct support.

5. A Gerschekronian Interpretation of NIE Developments

5.1 The Asian Experience Within a Gerschenkronian Framework
To interpret the Asian development experience within a Gerschenkronian framework,
a useful starting point is Gershenkon’s core arguments with respect to European
industrialization:
… the industrial development of Europe appears not as a series of mere
repetitions of the “first” industrialization but as an orderly system of gradu-
ated deviations from that industrialization
(Gershenkon, 1962, p. 44)
In addition:
European history should be seen as a pattern of substitution governed by the
prevailing—and changing—degree of backwardness
(Gershenkon, 1962, p. 359)
and that
… the higher the degree of backwardness, the more discontinuous the devel-
opment is likely to be
(Gershenkon, 1962, p. 45)
It would seem that the East and South East Asian experiences, in general, conform to
this view of “graduated deviation”. What one witnesses is the first developer of the
region, Japan, imitated, but with significant differences, by South Korea and the other
three “tigers”, then followed by less-developed countries. In the case of China one can
identify a whole mix of strategies borrowing from each of the other countries (FDI,
OEM, OBM, state enterprise, joint ventures and so on). However, there appears to be
 Innovation in Asian Industrialization 307
more uniformity and less policy innovation within the group of South East Asian NIEs,
where each country followed remarkably similar patterns of FDI-led development,
distinguished more by their time of entry, sector specialization and level of economic
development, than innovation in policy, institutions or industrial structure. By contrast,
the East Asian group revealed less uniformity and more distinctive strategic innovation.
This suggests that a key research task, from a Gerschenkronian perspective, is to
compare systematically the degree of innovation with the extent of imitation among
countries, and to identify reasons for divergence and similarities. While it is outside the
scope of this paper to undertake this task in any depth, it would appear that both
innovation and imitation exist side by side in the NIEs, and that observed divergences
are determined by initial starting conditions (in terms of resources and capabilities), the
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

timing of entry, and the nature and extent of international opportunity (especially in
terms of export markets and foreign investment) open to individual countries. In some
cases, a degree of strategic innovation may have been forced upon individual countries,
given the nature of existing resources. In others, notably within South East Asia, a
higher degree of imitation was consonant with available capabilities. Following the lead
of Singapore, the “path of least resistance” was for other South East Asian economies
also to invite foreign TNCs to lead export development by exploiting the abundant,
flexible, low-cost labour available.

5.2 Policy-making for Development: Substituting for Missing Prerequisites

Although it is reasonable to conclude that the development experience of East and
South East Asia broadly conform to Gerschenkron’s view, what is less clear are the
policy implications or lessons for other developing countries. To tackle this issue, it is
helpful to draw upon Gerschenkron’s concept of “substitution of missing prerequisites”
mentioned above. Directly contradicting the popular thesis of Rostow (1960), Ger-
schenkron argued that it was not sensible to invest in a standard set of prerequisites but,
on the contrary, it was necessary to substitute for missing prerequisites as defined by the
experience of some leading country or earlier, successful industrializer. Gerschenkron
also argued that this pattern of substitution was governed by the prevailing, and
changing, degree of backwardness. However, as far as translating this analytical method
or insight into development lessons or policies, Gerschenkon was very cautious.
Regarding actual deliberate policies for development he argued:

There is no intention to suggest that backward countries necessarily engaged

in deliberate acts of “substitution” for something that had been in evidence in
more advanced countries. Men in a less developed country may have simply
groped for and found solutions that were consonant with the existing condi-
tions of backwardness.
(Gerschenkron, 1962, p. 359)

In other words, knowing that some form of difference (or innovation) in strategy is
needed because of the experience of, and the altered circumstances caused by, earlier
industrializers, does not mean that a development path then becomes clear. This
implies that the process of strategic or policy innovation embodies a high degree of
experimentation and learning to cope with conditions of uncertainty and change.23
Under these circumstances, elaborate and detailed development plans, however well
articulated, are unlikely to generate development, except in the rare case that plans
exactly match unfolding future circumstances. Instead, simplicity and flexibility in
 308 M. Hobday

development strategy are required in order to respond to new circumstances and exploit
emerging opportunities as and when they arise.
In addition, Gerschenkron pointed out that, in many cases, the later industrializers
were not even conscious of the need to formulate a new strategy in response to changes,
but instead learned by trial-and-error, perhaps consciously or unconsciously learning
from some of the experiences of earlier industrializers.
Furthermore, while admitting that imitation did play a part in industrial develop-
ment, he also argued that:
the very concept of substitution is premised upon creative innovating activity,
that is to say, upon something that is inherently unpredictable with the help
of our normal apparatus of research.
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

(Gerschenkron, 1962, pp. 359–360)

To the extent that Gerschenkron was correct, and that the substitution of preconditions
is necessary for development, then latecomer development is essentially an innovative
activity. This implies that strategic innovation, as defined in the Introduction, is at the
heart of economic development. However, this may be taking the argument too far.
Indeed, as shown above, individual East and South East Asia countries relied on a mix
of strategic innovation and imitation as they moved through their various, endogenous,
stages of development.
Returning to the experience of the Asian NIEs, a key Gerschenkronian research
question is to what extent, and in which specific ways, did these latecomers substitute for
the missing prerequisites as exemplified by an arbitrary developed country (e.g. Japan
or the USA)? Table 4 presents some of these substitution mechanisms. The table shows
that in South Korea, state promotion of the embryonic chaebol in the early stages of
industrialization substituted for missing entrepreneurial, business and managerial ca-
pacity, evident in the USA and Japan (Jones & Sakong, 1980; Amsden, 1989). In the
case of Singapore and Malaysia, the TNC subsidiaries substituted for a lack of export
experience and technological capability. These were promoted through various subsi-
dies, export promotion schemes, tax holidays and infrastructural support. As Section 3
showed in the cases of Taiwan, South Korea and Hong Kong, the OEM system
substituted for direct access to advanced markets and provided a substitute for missing
technological capability in the form of a “technological training school”.
In each Asian case, the concentrated focus on export-led growth substituted for
small domestic markets, especially in the early stages of development of electronics.
Regarding technology, in each case foreign technology transfer and local improvements
substituted for the absence of domestic engineering capacity and R&D resources.
In the case of Thailand, Malaysia and Vietnam, in line with Gerschenkron’s
argument, we see evidence of further graduated deviations, with each country ap-
proaching electronics in a familiar FDI-led manner. China, another late entrant, in
some ways more backward than the four leading NIEs and the leading South East Asian
nations, is following with a range of strategies and paths that both imitate and build
upon both East and South East Asian models of growth.
Although, as Gerschenkron insisted, direct lesson-making is inappropriate, the
process of identifying the substitutes for missing prerequisites could potentially play a
useful role in assisting other developing countries to learn lessons from East and South
East Asia and in developing their own distinctive strategies for development. Each
developing country embodies it own particular advantages and disadvantages, shaped
by its stage of backwardness. By observing the mechanisms and extent of strategic
innovation in the Asian NIEs, other developing countries might find new ways of
 Innovation in Asian Industrialization 309
Table 4. Substitutes for missing prerequites in Asian NIEs

Prerequisite for growth—evident in the Asian substitution for missing

USA or Japan prerequisite

Strong entrepreneurial and managerial Korea—state sponsorship of the chaebol

capacity Singapore—subsidies for TNC
subsidiaries

Large internal markets/access to export Korea, Taiwan, Hong Kong—OEM

markets via FDI/trade subcontracting system—export-led
growth
South East Asia—exports via TNC
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

subsidiaries
Access to advanced technology/R&D Singapore/Malaysia/Thailand—foreign
clusters direct investment—TNC subsidiaries
Korea, Taiwan, Hong Kong, OEM, Joint
ventures

overcoming disadvantages, including the lack the resources and capabilities needed for
development. Each nation has its own set of missing prerequisites and (hopefully)
distinctive methods for substituting for them. Beyond this, it is not possible to draw
universal or specific lessons for other developing countries. Each country will need to
draw its own lessons, based not only on its particular conditions but also the changing
external circumstances, in part caused by the Asian NIEs themselves (e.g. the strong
attraction of FDI to Asia and the success of the NIEs in entering the electronics
industry).

5.3 Interpeting Gershenkron in the Modern Context

Although Gershenkron’s approach provides useful insights into modern Asian develop-
ment, it is also important to emphasize that his analysis was concerned with an early
period of European industrial history, rather than modern day Asian growth, and
circumstances have changed considerably. In at least three areas, modern circum-
stances point to new challenges for the developing countries of today.
First, in contrast with early European industrialization, today, TNCs can and do
play an important role in industrial development. FDI has expanded and deepened
considerably in the past 20 years or so, leading some to include FDI in definitions of
“globalization”. This factor was not present to the same extent in early European
industrialization and, as we have seen, in the case of South East Asia provided a
valuable substitute for local exporting enterprise in leading export development.
Second, there was a heavy bias in Gerschenkron’s work on static economies of scale,
appropriate for the large-scale, capital-intensive production of steel and the other heavy
industries central to 19th Century industrialization. Today, in many leading sectors the
prerequisites for industrialization have changed. Electronics and software, for example,
are light industries that lend themselves to an international division of labour in which
low-cost skilled labour can be an important initial entry point. Knowledge accumula-
tion and catching up in these types of industries are as much to do with management
ingenuity, engineering and product design capabilities as traditional production scale
economies.
Third, Gerschenkron’s work did not encompass the effects of major changes in the
international technology environment and, in particular, what Perez and Freeman call
 310 M. Hobday

“techno-economic paradigms” (Perez, 1983, 1985; Freeman, 1991, 1993; Freeman &
Louca, 2001). Indeed, Gerschenkron (1962, p. 364) recognized this as a limitation in
his own work, and suggested that research into new technological innovations in the
advanced countries was essential for understanding the prospects facing contemporary
latecomers.

6. Conclusion: Lessons for other Developing Countries?

Gerschenkron poses a very difficult challenge for those wishing to draw lessons from
successful industrializers such as the Asian NIEs. His interpretation would suggest that
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

there are few direct lessons from East and South East Asia for other countries and
certainly no transferable ready-made model of development, either for electronics as a
leading sector, or for economic strategy more broadly. Export-led growth may not be
an option for all and electronics may now be too crowded for many others to follow,
especially with the entry of China and Mexico as major exporters. Other national
development paths and alternative leading sectors will need to be identified and
followed (e.g. the software industry in Bangalore India) which match and exploit the
particular capabilities of specific countries. In addition, with any development strategy,
innovation and trial-and-error learning will be required, with little predictability in
advance and no guarantee of success.
The experience of the NIEs suggests that today’s latecomers need to draw upon
their own distinctive resources which will, in part at least, be shaped by their prevailing
stage of backwardness. For example, the OEM system of East Asia did not simply
“arrive from nowhere”. It was created endogenously, in collaboration with overseas
buyers in the 1960s, then nurtured, invested in and greatly expanded in subsequent
decades. Similarly, the TNC subsidiary-led growth of South East Asia, on the scale
witnessed, did not simply “occur” but arose as another institutional innovation. Like
OEM, FDI was systematically attracted, supported and rewarded. As a result, it
became progressively embedded in the industrial and technological infrastructure of
South East Asia, enabling technology transfer and export market entry on a very large
scale.
As Gerschenkron pointed out, no policy model based on the experience of others
can automatically succeed and no mechanistic development formula is likely to produce
industrialization. Similarly, the lesson of East and South East Asia is that a very
demanding “climb up” process is required, based on a mixture of institutional, policy
and technological innovation and careful imitation of certain aspects of other more
successful countries in similar circumstances.
To the extent that development stages are created, not merely passed through, they
are essentially an endogenous, innovative activity. This applies both to early developers
who create new paths and to latecomers who respond, adapt and act upon the new
conditions facing them. Imitation alone is likely to be insufficient to produce catch-up
development. The Asian NIEs reveal a wide variety of development paths and exper-
imentation not only in terms of government policy and industry structure but also in
patterns of industrial ownership, firm size and mechanisms for acquiring technology
and entering export markets.
The lesson from the observed “graduated deviation” of NIE development paths is
that today’s developing countries also need to find ways of substituting for the missing
prerequisites of industrialization. Each individual developing country will need to seek
out its own development path based on its own distinctive resources. This, in turn, will
 Innovation in Asian Industrialization 311
require them to develop the capabilities to enact the strategic innovation needed to
exploit market and technological opportunities.

Notes
1. This paper uses the term “strategic innovation” or “policy innovation” to describe this broad
category of innovation and to distinguish it from the narrower concept of “technological
innovation”. While technological transfer and innovation is often discussed in studies of
economic development (Fransman & King, 1984; Kim, 1997), innovation in its broader
strategic sense has received less attention. Strategic innovation can be defined as the
introduction and implementation of new ideas, new institutions, new governmental policies
and experimental business strategies.
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

2. The paper does not return to the early debates over linear “stages of development” models
popularized by Rostow (1960). Instead, it accepts the arguments put forward by Kuznets
(1954, 1966) and others that there is little evidence that linear, repeatable stages can be used
to explain development or draw lessons for today’s developing countries. However, the
notion of historical stages as a limited and partial device for summarizing differences, as well
as similarities, in certain aspects of development, is used in this paper in line with Ger-
schenkron’s approach to European history. Today, this device would be called
“benchmarking”.
3. Many studies draw direct lessons from the Asian experience with little regard to differences
between the Asian economies or the specific needs of particular “receptor” countries. For a
review of these studies see Mbula (2002).
4. Gerschenkron’s stages were not about short-run periods of rapid growth or take-off
specifically or, indeed, individual sectors. They were mostly concerned with the great sweeps
of history, sometimes covering more than a century. However, as the paper shows, the
specific stages of the East and South East Asian take-off in electronics are amenable to a
Gerschenkronian interpretation in that they reveal both differences and similarities in
broader patterns of economic development across countries and represent a great “spurt” of
industrialization.5. See his major text: Economic Backwardness in Historical Perspective: A
Book of Essays, published by The Belknap Press of Harvard University Press Gerschenkron,
1962).
6. For a recent summary in relation to the Asian NIEs, see Whestphal (2002).
7. Section 5 interprets this evidence within the debate outlined above.
8. Firm-level growth rates were much faster, sometimes in the region of 30% per annum for
fairly prolonged periods.
9. Singapore is treated with other South East Asian NIEs in Section 4.
10. Note that other forms of technology acquisition were also important (and sometimes linked
to OEM), including licensing, joint ventures and other forms of subcontracting, as well as
FDI in the early stages of export growth.
11. Cyhn (2002) for an in-depth analysis of the OEM system in Korea; also see Hobday (1995),
which compares the OEM in South Korea with Taiwan and Hong Kong.
12. For example, see the typical “MIT” models used by authors such as Abernathy & Clark
(1985) and Clark & Fujimoto (1991).
13. See Hwang (2000) and Lee (2001), respectively.
14. See Chaponniere & Fouqin (1989) for the case of electronics; see Dahlman & Sananikone
(1990) for industry in general; Schive (1990) shows that one should not overlook the
important role of foreign firms via FDI in the very early stages industry start up in Taiwan;
Bloom (1990, 1991) shows an equivalent role for FDI in Korea’s early start in electronics.
15. Note that recently this situation has changed, with a huge influx of FDI into Korea. During
the 4-year period (1998–2001) after the crisis, Korea attracted around US$52 billion in FDI,
which is more than double the entire amount of the previous four decades.
16. For Taiwan, see Chaponniere & Fouquin (1989). Schive (1990) shows how Taiwanese
SMEs attracted more TNCs into Taiwan in the early stages, creating a series of backward
and forward linkages; for Hong Kong’s experience see Berger & Lester (1997).
17. In fact, electronics contributed around 34% to gross manufacturing value in the early 1990s
(EDB, 1992, p. 21).
18. For stages in general in Thai industry, see Intarakumnerd and Virasa (2002); for firm-level
 312 M. Hobday

development stages, see Chairatana (1997); for electronics in Thailand, see Poapongsakorn
& Tonguthai (1998).
19. See Ca & Anh (1998) for Vietnam and Thee & Pangestu (1998) for Indonesia.
20. See Ngoh (1994) for the case of Motorola and Lim (1991) for the case of Intel.
21. It is also very interesting to note that China has more recently adopted both OEM and FDI
strategies in unison, along with the other strategies of previous industrializers (see Section 5).
22. See Guyton (1994) for a critique of Japanese corporate strategies in Malaysia. Bartlett &
Ghoshal (1987a, b) assessed decentralization and the importance of US TNC subsidiary
decision-making.
23. In the field of organizational development this point has been made forcibly by authors such
as Lindblom (1959), who describe the practice of “muddling through” under conditions of
uncertainty. Also see similarly arguments by Klein & Meckling (1958) for decision-making
in development projects.
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

References
Abernathy, W.J. & Clark, K.B. (1985) Innovation: mapping the winds of creative destruction,
Research Policy, 14, pp. 3–22.
Amsden, A. (1989) Asia’s Next Giant: South Korea and Late Industrialisation (New York, Oxford
University Press).
Ariffin, N. & Bell, M. (1998) Firms, politics and political economy: patterns of subsidiary–parent
linkages and technological capability-building in electronics TNC subsidiaries in Malaysia, in:
K.S. Jomo, G. Felker & R. Rasiah (Eds) Industrial Technology Development in Malaysia
(London, Routledge).
Bartlett, C. & Ghoshal, S. (1987a) Managing across borders: new strategic requirements, Sloan
Management Review, 28, pp. 7–17.
Bartlett, C. & Ghoshal, S. (1987a) Managing across borders: new organisational requirements,
Sloan Management Review, 29, pp. 43–53.
Bell, M., Hobday, M., Abdullah, S., Ariffin, N. & Malik, J. (1996) Aiming for 2020: A
Demand-driven Perspective on Industrial Technology Policy in Malaysia, Final Report to Ministry
of Science, Technology and Environment (Malaysia), World Bank/UNDP.
Berger, S. & Lester, R.K. (Eds) (1997) Made by Hong Kong (Hong Kong, Oxford University
Press).
Bloom, M. (1991) Globalisation and the Korean electronics industry, EASMA Conference, The
Global Competitiveness of Asian and European Firms, Fontainbleau, 17–19 October.
Bloom, M. (1992) Technological Change in the Korean Electronics Industry (Paris, Development
Centre Studies, OECD).
Ca, T.N. & Anh, L.D. (1998) Technological dynamism and R&D in the export of manufactures
from Vietnam, in: D. Ernst; T. Ganiatsos & L. Mytelka (Eds) Technological Capabilities and
Export Success in Asia (London, Routledge).
Chairatana, P. (1997) Latecomer catch-up strategies in the semiconductor business: the case of
Alphatec Group of Thailand and Anam Group of Korea, MSc Thesis, SPRU, University of
Sussex.
Chaponniere, J.R. & Fouquin, M. (1989) Technological Change and the Electronics Sector—Perspec-
tives and Policy Options for Taiwan, Report Prepared for OECD Development Centre Project,
May (Paris, OECD).
Chenery, H.H. (1968) The role of industrialization in development programmes, in: A.N.
Agarwala & S.P. Singh (Eds) The Economics of Underdevelopment (Oxford, Oxford University
Press).
Chou, T.C. (1992) The experience of SMEs’ development in Taiwan: high export-contribution
and export-intensity, Rivista Internazionale di Scienze Economiche e Commerciali, 39, pp. 1067–
1084.
Clark, K.B. & Fujimoto, T. (1991) Product Development Performance: Strategy, Organization and
Management in the World Auto Industry (Boston, MA, Harvard Business School).
Cyhn, J.W. (2002) Technology Transfer and International Production: The Development of the
Electronics Industry in Korea (Cheltenham, Edward Elgar).
Dahlman, C.J. & Sananikone, O. (1990) Technology Strategy in the Economy of Taiwan: Exploiting
Foreign Linkages and Investing in Local Capability (Washington, World Bank).
EDB (1992) The Electronics Industry in Singapore (Singapore, Economic Development Board).
 Innovation in Asian Industrialization 313
Fransman, M. & King, K. (Eds) (1984) Technological Capability in the Third World (London,
Macmillan).
Freeman, C. (1991) Innovation, changes of techno-economic paradigm and biological analogies
in economics, Revue Economique, 42, pp. 211–232.
Freeman, C. (1993) Technological revolutions and catching-up: ICT and the NICs, in: J.
Fagerberg, B. Verspagen & N. von Tunzelmann (Eds) The Dynamics of Technology, Trade and
Growth (Aldershot, Edward Elgar).
Freeman, C. & Louca, F. (2001) As Time Goes By: From the Industrial Revolution to the Information
Revolution (Oxford, Oxford University Press).
Gerschenkron, A. (1962) Economic Backwardness in Historical Perspective (Cambridge, MA, Har-
vard University Press).
Gerschenkron, A. (1963) The early phases of industrialisation in Russia, in: W.W. Rostow (Ed.)
The Economics of Take-off into Sustained Growth (London, Macmillan).
Guyton, L.E. (1994) Japanese FDI and the Transfer of Japanese Consumer Electronics Production to
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

Malaysia, Report Prepared for UNDP, K.L., Malaysia.

Hobday, M. (1994) Export-led technology development in the four dragons: the case of electron-
ics, Development and Change, 25, pp. 333–361.
Hobday, M. (1995) Innovation in East Asia: The Challenge to Japan (London, Edward Elgar).
Hong, S.G. (1997) The Political Economy of Industrial Policy in East Asia: The Semiconductor
Industry in Taiwan and South Korea (Cheltenham, Edward Elgar).
Hwang, C.Y. (2000) The aircraft industry in a latecomer economy: the case of South Korea,
DPhil Thesis, Brighton, SPRU, University of Sussex.
Hwang, H.R. (1998) Organisational capabilities and organisational rigidities of Korean Chaebol:
case studies of semiconductor (DRAM) and personal computer (PC) products, DPhil Thesis,
SPRU, University of Sussex.
Intarakumnerd, P & Virasa, T. (2002) Taxonomy of Government Policies and Measures in Supporting
Technological Capability Development of Latecomer Firms, Science, Technology and Innovation
Policy Research Department, National Science and Technology Development Agency
(NSTDA), Thailand and College of Management, Mahidol University, Thailand, NSTDA
Working Paper.
James, W. (1990) Basic Directions and Areas for Cooperation: Structural Issues of the Asia-Pacific
Economies, Asia Pacific Cooperation Forum, Session 2, 21–22 June, Korea Institute for
International Economic Policy, Seoul.
Johnstone, B. (1989) Taiwan holds its lead. local makers move into new systems, Far Eastern
Economic Review, 31 August, pp. 50–51.
Jones, L.P. & Sakong, I.L. (1980) Government, Business, and Entrepreneurship in Economic Develop-
ment: The Korean Case, Harvard East Asian Monographs No. 91 (Cambridge, MA, Harvard
University Press).
Kim, L. (1997) Imitation to Innovation: The Dynamics of Korea’s Technological Learning (Boston,
MA, Harvard Business School Press).
Klein, B. & Meckling, W. (1958) Application of operations research to development decisions,
Operations Research, May–June, pp. 352–363.
Kuznets, S. (1954) Underdeveloped countries and the pre-industrial phase in the advanced
countries in United Nations, Proceedings of the World Population Conference, Vol. 5, New York.
Kuznets, S. (1966) Modern Economic Growth: rate, structure, spread (New Haven, Yale University
Press).
Lee, T.J. (2001) Technological capabilities and international relations in developing countries:
case studies of the nuclear fuel cycle in South Korea, DPhil Thesis, Brighton: University of
Sussex.
Lim, P. (1991) Steel: From Ashes Rebuilt to Manufacturing Excellence (Pataling Jaya, Malaysia
Pelanduk Publications).
Lindblom, C.E. (1959) The science of ‘’muddling through”, Public Administration Review, 19,
pp. 79–88.
Mathews, J.A. & Cho, D.S. (2000) Tiger Technology: The Creation of a Semiconductor Industry in
East Asia (Cambridge, Cambridge University Press).
Mbula, E.K. (2002) Building capabilities: Can Sub-Saharan countries learn from the East Asian
experience?, MSc Thesis, SPRU, University of Sussex.
Ngoh, C.L. (1994) Motorola Globalisation: The Penang Journey (K.L. Malaysia, Lee and Sons).
Perez, C. (1983) Structural change and the assimilation of new technologies in the economic and
social system, Futures, 15, pp. 357–375.
 314 M. Hobday

Perez, C. (1985) Micro-electronics, long waves and world structural change, World Development,
13, pp. 441–463.
Poapongsakorn, N. & Tonguthai, P. (1998) Technological capability building and the sustainabil-
ity of export success in Thailand’s textile and electronics industries, in: D. Ernst; T. Ganiatsos
& L. Mytelka, (Eds) Technological Capabilities and Export Success in Asia (London, Routledge).
Rasiah, R. (1994) Flexible production systems and local machine tool sub-contracting: electronics
components transnationals in Malaysia, CambridgeJournal of Economics, 18, pp. 279–298.
Rostow, W.W. (1960) The Stages of Economic Growth: A Non-Communist Manifesto (Cambridge,
Cambridge University Press).
Schive, C. (1990) The Foreign Factor: The Multinational Corporation’s Contribution to the Economic
Modernisation of the Republic of China (Stanford, CA, Hoover Institution Press).
Thee, K.W. & Pangestu, M. (1998) Technological capabilities and Indonesia’s manufactured
exports, in: D. Ernst, T. Ganiatsos & L. Mytelka (Eds) Technological Capabilities and Export
Success in Asia (London, Routledge).
Downloaded By: [Middle East Technical University] At: 23:03 25 February 2007

Wade, R. (1990) Governing the Market: Economic Theory and the Role of Government in East Asian
Industrialisation (Princeton, NJ, Princeton University Press).
Whestphal, L.E. (2002) Technology strategies for economic development in a fast changing
global economy, Economics of Innovation and New Technology, 11, pp. 277–320.
Wong, P.K. (1992) Technological development through sub-contracting linkages: evidence from
Singapore, Scandinavian International Business Review, 1, pp. 28–40.
Wong, P.K. (1998) Patterns of technology acquisition by manufacturing firms in Singapore,
Singapore Management Review, 20, pp. 43–64.
World Bank (1993) The East Asian Miracle: Economic Growth and Public Policy (World Bank, New
York, Oxford University Press).
Yue, C.S. (1985) The role of foreign trade and investment in the development of Singapore, in:
W. Galenson, (Ed.) Foreign Trade and Investment: Economic Development in the Newly Industrial-
ising Asian Countries (Wisconsin, University of Wisconsin Press).