Wider Research Paper 2009/50 Technological Innovation ...
Research Paper No. 2009/50
Technological Innovation,
Entrepreneurship and Development
David B. Audretsch and Mark Sanders*
November 2009
Abstract
Industrialization has long been seen as the answer to underdevelopment and poverty.
First this led countries to follow protectionist import substitution policies but as these
failed developing countries have opened up to trade and FDI and tried to follow
strategies of export driven industrialization. If we consider the share of non-OECD
countries in global trade in manufactures, this has been a big success. But has it?
Developed countries still retain their competitive advantage in the innovative and fast
growing industries of the future and for every success story in Asia there are at least two
tales of woo in Africa. In this paper we present a two region product life cycle model of
global specialization and trade. In it we analyse the impact of three major shocks to the
gradually globalizing and integrating world economy and show that these shocks have
caused a transition in the global specialization pattern. The advanced and previously
industrialized countries have arguable made the transition to an entrepreneurial
economy in which innovation, creativity and high value added in early stage activity…./
Keywords: globalization, innovation, trade, development
JEL Codes:
F01, J31, O1, P0
Copyright © UNU-WIDER 2009
* Max Planck Institute of Economics, Research Group on Entrepreneurship, Economic Growth and
Public Policy, Jena, Germany. Email of corresponding author: m.w.j.l.sanders@uu.nl
This paper was presented at the UNU-WIDER and UNU-MERIT Research Workshop on
Entrepreneurship, Technological Innovation, and Development, held in Maastricht, the Netherlands,
30–31 October 2008. This paper is also published as UNU-MERIT Working Paper 2009-052.
UNU-WIDER gratefully acknowledges the financial contributions to the project by the Finnish Ministry
for Foreign Affairs, and the financial contributions to the research programme by the governments of
Denmark (Royal Ministry of Foreign Affairs), Finland (Finnish Ministry for Foreign Affairs), Sweden
(Swedish International Development Cooperation Agency—Sida) and the United Kingdom (Department
for International Development).
ISSN 1810-2611
ISBN 978-92-9230-229-0
are the basis of competitive advantage, whereas the developing world by-and-large has
absorbed mature industrial activities based on the Heckscher–Ohlinian competitive
advantage based on cheap unskilled labour. The key exogenous shocks that have led to
this shift are the collapse of communism, the introduction of information and
communication technologies and the opening up of large, populous developing
countries such as India, China, and Brazil. Our model predictions are very much in line
with observed trends in developing and developed economies and as such provides
insights in the possible underlying mechanisms at work.
Abbreviations
CES
Constant Elasticity of Substitution
FDI
Foreign direct investment
ICT
Information and communication technology
GFCF
Gross fixed capital formation
GPT
General
purpose
technology
OECD
Organisation for Economic Cooperation and Development
R&D
Research and development
The World Institute for Development Economics Research (WIDER) was
established by the United Nations University (UNU) as its first research and
training centre and started work in Helsinki, Finland in 1985. The Institute
undertakes applied research and policy analysis on structural changes
affecting the developing and transitional economies, provides a forum for the
advocacy of policies leading to robust, equitable and environmentally
sustainable growth, and promotes capacity strengthening and training in the
field of economic and social policy making. Work is carried out by staff
researchers and visiting scholars in Helsinki and through networks of
collaborating scholars and institutions around the world.
www.wider.unu.edu publications@wider.unu.edu
UNU World Institute for Development Economics Research (UNU-WIDER)
Katajanokanlaituri 6 B, 00160 Helsinki, Finland
Typescript prepared by Lisa Winkler at UNU-WIDER
The views expressed in this publication are those of the author(s). Publication does not imply
endorsement by the Institute or the United Nations University, nor by the programme/project sponsors, of
any of the views expressed.
1 Introduction
Globalization and technical change have been identified as key sources of structural
change in industrialized and developing countries alike (e.g. Grossman and Helpman
1991; Helpman 1998). At the same time both trends can be understood as the outcome
of economic processes rather than exogenous to the global economic system. That begs
the question what fundamental economic mechanisms may underlie the interaction
between globalization, technical innovation, and structural change. The purpose of this
paper is to address that question. We do so by analysing the impact of three important
shocks to the global economy that have occurred since the 1980s.
Arguably, the fall of communism in essence caused an exogenous global reduction in
political risks and allowed large low skilled labour abundant economies to join the
global economy (Audretsch 2007). This opening up of opportunities for economic
integration coincided with the, also exogenous, introduction of a new general purpose
technology (GPT) in the form of information and communication technologies (ICT)
over the past 25 years (Helpman 1998).
In this paper we argue that this combination of shocks caused the developed economies
to make the transition from an industrial to an entrepreneurial economy in response to
new opportunities; moreover, the entrepreneurs in the developed countries changed the
pattern of national and international specialization in such a way that the developed
countries experience wage divergence among skill groups and increase the share of
new, knowledge-intensive products and services in their economy. This corresponds
with observed transition to the ‘knowledge economy’ and the ‘rise of the creative class’
that has been well established in the literature (Florida 2003).
Through outsourcing and FDI the developing and emerging economies on the other
hand increasingly engage in the production of mature industrial goods (manufactures)
and compete on the basis of their (static) comparative advantage, that is low wages.
These countries enter the industrial era, but based on technology transfer that is largely
initiated by international entrepreneurs.
It is quite clear from the empirical literature (e.g. Keller 1996) that not all countries
make that transition equally successfully and that for example absorptive capacity is a
key in attracting and benefiting from the existing international entrepreneurship. In
addition, local entrepreneurs and firms in developing and emerging economies also
seize new opportunities and engage in export (one might call that ‘insourcing’) or even
initiate FDI themselves. Moreover, as countries develop, they can be expected to move
from agricultural self-sufficiency with abundant low skilled labour, to industrial export-
oriented with increasing levels of human capital to an entrepreneurial and services-
oriented mature economy with skilled labour and sophisticated demand. In this paper,
however, we choose to limit ourselves to analysing the world as if it consisted of two
regions. One advanced region that produces advanced early stage innovative products
and one developing region that produces routine, mature standardized products. We like
to stress at this point that this simplification is not intended to be a statement about the
world as it is but rather is a useful abstraction that allows us to zoom in on the
mechanism we are primarily interested in.
1
To outline our basic argument we build on elements from life cycle trade models as in
Krugman (1979) and Grossman and Helpman (1991), models of endogenous skill
biased technical change as in Acemoglu (2002b) and Sanders (2005) and notions from
the large literature on the entrepreneurial function surveyed by for example Audretsch
and Acs (2005).
Our model shows how globalization and technological innovation drive the structural
shift towards the entrepreneurial economy in the advanced region, but entrepreneurs in
turn may drive the process of globalization and GPT exploration, both in the advanced
and developing economy. We conclude from our analysis that the opening up and
exploitation of opportunities for trade and GPT related products and services may have
hurt especially the low skilled workers in advanced economies, but globalization
through reduced political risks in the emerging countries unambiguously benefited all
workers in both regions.
In the remainder of this paper we first offer the data and evidence from the literature
that underlie the stylized facts we intend to explain: the rise of an industrial economy in
(parts of) the developing world, the shifting patterns of international specialization, the
shift in aggregate labour demand in the advanced countries and finally the rise of an
entrepreneurial economy there. This section also presents the evidence in support of
some key assumptions in our model, notably the product life cycle and the role of
international entrepreneurs in actively relocating mature production activities to
developing countries. Section 2 also establishes the coincidence of exogenous shocks
we offer as the catalyst for these developments. Section 3 presents our arguments in a
formal model, Section 4 analyses the implications and concludes.
2 Stylized
facts
2.1 Political risk, global labour supply, and a new general purpose technology
Globalization and the advent of new ICT technology cannot reasonably be considered in
isolation. It is obvious that the spectacular drop in the price of communication has been
instrumental in allowing firms to expand their operations globally. It is also more than
reasonable to assume that the development of ICT technology has benefited greatly
from the boom in international demand for goods and services in general.
Globalization, however, would not have occurred to the degree that it has if the
fundamental changes were restricted to the advent of the microprocessor and
telecommunications. During the post-war era most trade and economic investment was
confined to Europe and North America, and later a few of the Asian countries,
principally Japan and the Asian Tigers. Technology alone could not have overthrown
that, largely political, constellation. Trade with countries behind the iron curtain was
restricted and in some cases prohibited. Even trade with Japan and other Asian and
South American countries was highly regulated. Similarly, investments in politically
unstable countries resulted in episodes of national takeovers and confiscation where the
foreign investors lost all or part of their investments.
We therefore argue that it took a political revolution in significant parts of the world to
reap the full benefits of these technological changes. The political counterpart of the
technological revolution was the increase in democracy and concomitant stability in
2
areas of the world that had previously been inaccessible. As Thurow (2002: 25–26)
pointed out, ‘Much of the world is throwing away its communist or socialist inheritance
and moving towards capitalism. Communism has been abandoned as unworkable
(China), imploded (the USSR), or has been overthrown (Central and Eastern Europe)’.
Figure 1 illustrates this global political revolution by showing countries switching from
one party (light shaded) or communist regimes (medium shaded) to liberal democracies
(dark shaded) between 1950 and 2000.
The ‘victory’ of democracy and capitalism over communism also provoked a renewed
commitment to the principles of free trade and international competition in countries
such as India and Brazil. India, for example, became accessible as a trading and
investment partner after opening its economy in the early 1990s. With some of these
areas (re)joining the world economy for the first time in decades, the post-war
equilibrium came to a sudden end.
What most distinguished the new players in the global economy was their relative
abundance in cheap labour with comparatively low levels of education and productivity.
For example, in the early part of the 1990s, the daily earnings of labour were estimated
to be US$92.24 in the United States and US$78.34 in the European Union. This stood in
Figure 1: The Rise of democracy
3
Figure 2: Share in global active population in 2005
Source: ILO (2005). Available at: http://www.ilo.org/public/english/bureau/stat/portal/index.html
sharp contrast to wages of only US$6.14 in Poland and US$6.45 in the Czech Republic
shortly after the Berlin Wall fell. With Asia, the wage gap was even greater, as daily
earnings were USUS$1.53 in China, USUS$2.46 in India and USUS$1.25 in Sri Lanka.
Where previously trade barriers and political control kept their economies disconnected
and allowed such disparity to persist, China and India alone now account for over one
third of the world’s active population.
We thus establish Fact 1: Globalization is characterized by the entry of large
populous low wage economies
The potential labour force in countries like China, with 1.3 billion and India with 1.1
billion inhabitants dwarfs the population and workforce in North America and Europe.
The global economic system thus had to deal with a sudden expansion of the supply of
cheap manual labour. In any market economy the response to such a shock is
predictable. In the incumbent players’ economies the wages decline for the jobs such
workers can do and rise in complementary occupations (including the production of
non-tradables). In addition, following standard Heckscher–Ohlin trade theory, one
expects a shift in international trade patterns and increasing standards of living in the
new entrant countries. The data over the past 10 years show all these trends.
But globalization is certainly not all ‘bad news’ for the North. The political revolution
created opportunities for entrepreneurs associated with the gaping labour cost
differences and the emerging markets for their products. As long as the Wall stood, and
countries such as China and Vietnam remained closed, the large discrepancies in wage
rates could be maintained without eliciting responses in trade and FDI.
Trade with and certainly direct investment by foreign companies was simply prohibited
by local governments or considered to be too risky. With opening up to trade and FDI
4
Figure 3: FDI outflow in % GDP
Source: UNCTAD (2003).
Figure 4: FDI inflow in % GDP
Source: UNCTAD (2003).
5
Figure 5: Returns on US foreign assets
Source: Data taken from Lane and Milesi-Feretti (2003), figure 10: 31.
these countries allow Northern entrepreneurs to step in and arbitrage away such
differentials by outsourcing and off-shoring, making good money in the process.1
The political shift also had a huge impact on global business opportunities. As can be
verified in Figures 3 and 4 volumes of FDI increased rapidly between 1990 and 2000.
Meanwhile one can verify in Figure 5 that the rate of return and yield on foreign assets
fell between 1983 and 2001 (by 8 and 3 basis points on average per year, respectively),
indicating a drop in risk premia as Western investors apparently were willing to invest
larger amounts at lower rates of return.
And we establish fact 2: Political risks declined worldwide
The Northern entrepreneurs were aided and facilitated in setting up such operations by
the recently developed information and communications technologies that virtually
destroyed distance and provoked serious reorganization of production in itself. This new
GPT predates the fall of the Wall but the North was still in the middle of exploring and
realizing its full potential when political events added an international dimension to the
exploration process and provided new and highly profitable opportunities for
application. It is well-established that ICT can be considered a GPT (in fact the concept
was developed with ICT in mind by Bresnahan and Trajtenberg 1995).
As Figure 6 illustrates, by 2000 most OECD countries have over 40 per cent of their
workforce using computers. Its rapid introduction in all aspects of the Northern way of
1 Outsourcing refers to placing parts of the production chain out of the firm and does not necessarily
cross international borders. Off-shoring refers to placing (parts of) the firms activities abroad but
maintaining control.
6
Figure 6: PCs per 100 white-collar workers
Source: OECD (2005).
life cannot have escaped anyone but the smallest infants and most senile senior citizens.
Obviously, by ‘destroying distance’ this technology was particularly effective in
reducing the costs of global operations.
This establishes the third fact: ICT technology is a pervasive GPT
We conclude that the drop in political risk, the expansion of the low skilled labour force
in the South and the introduction of a new GPT are the fundamental exogenous events
that triggered structural change in the last quarter of the 20th century. They all created
new opportunities for Northern entrepreneurs to the extent that they took a leading role
in shaping global production and trade. Their response to these new opportunities is
what effectuated the changes we set out to explain. Before proceeding with modelling
their role, however, we first establish those changes as empirically relevant facts in
North and South.
2.2 Shifting international specialization and the industrial economy in the South
(East)
A first indication of changing patterns of global economic activity is given by the more
traditional measures of transnational activity. These traditional measures include trade
(exports and imports), FDI (inward), international capital flows, and inter-country
labour mobility. The overall trend for all of these measures has been strongly positive.
Both exports and imports have become more important and accounted for a greater
share of economic activity. Total world exports increased from US$1.3 trillion in 1970
to nearly US$5 trillion in 2000, in constant dollars and exports have increased their
share in total output consistently since 1820 (Torres 2001: table 1), but as Figure 7
7
Figure 7: Share of exports in world GDP
Source: Data taken from Torres (2001), figure 1: 11.
shows also accelerates since the 60s.2 Exports matter absolutely, relatively and
increasingly.
Of course higher total exports could simply be more of the same products by the same
players crossing more borders. But this is not the case. The economic structure has
fundamentally changed over the past 25 years in the developing South. Where the South
previously relied on local production, agricultural, and natural resources exports while
importing most manufactures, they now increasingly compete for a share in the global
manufactures and even services markets. Especially in South-East Asia, where
land/labour ratios are low, manufacturing, not agriculture and/or mining is the key
sector of economic growth. Mayer and Wood (1999) predicted that the large economies
of China and India would grow towards manufacturing shares between 70–80 per cent
of exports and by the mid nineties these countries had already reached those levels.
China and India are by far the most impressive examples of rapid industrialization in the
region but as OECD (2005) shows the overall non-OECD share in world trade and
manufactures is increasing steadily.
A closer look at the data shows that it is especially intermediate industrial products from
South-East Asia that are responsible for this upward trend (see WTO 2005 and Feenstra
and Hanson 2004). In several countries in South-East Asia and some in South America,
industrial employment and output shares rose as a result of this outsourcing wave and
by the turn of the last century various industrialization processes have gained
momentum to the point that it is fair to speak of emerging industrial societies. In Taiwan
and South Korea for example, employment in agriculture has dropped to around 10 per
cent, a level comparable to that of New Zealand today and half that of the
2 Longer time series show a stagnant period in the Interbellum (1918–39) and First and Second World
War for well-known and obvious reasons. The export share in 1913 was 8.7 per cent and in 1950
stood at 7 per cent of global GDP.
8
Figure 8: FDI inflow in % GFCF
Source: UNCTAD (2003).
Netherlands in 1950. The large populous economies in the East are also rapidly
industrializing and with over 50 per cent of the workforce still in agriculture have a
huge industrial labour reserve yet to absorb.
Evidence presented by Antras (2005) suggests that Northern investors and entrepreneurs
played a vital role in fueling these waves of industrialization.3 International flows of
capital and in particular of FDI towards these new kids on the industrialized block have
exploded since the late 1980s and increasingly substituted for domestic savings as a
source of gross fixed capital formation (GFCF).
In general it is fair to say that the rise of industrial production in the South, at least in its
initial stages, was actively promoted by actions of economic agents from the North.
We thus establish fact 4: Northern Entrepreneurs fueled South-(East)-ern
industrialization
2.3 Shifting labour demand and the entrepreneurial economy in the North
In the North, meanwhile, traditional industrial employment in large industrial firms has
been falling as production was relocated. However, overall employment and output
have still grown, mainly due to small firm job creation in ICT related services. The
well-known transition towards the services economy is clearly illustrated in Figure 8.
3 Antras (2005) offers an incomplete contract model to explain why northern firms and entrepreneurs,
rather than southern imitators are the main initiator and vehicle of international technology transfer.
9
Figure 9: Employment shares in 17 OECD countries
Source: Data were taken from Iversen and Rosenbluth (2006), figure 4: 29.
In particular the widening gap between industry and services since 1970 indicates an
unprecedented labour reallocation. Arguably for the first time in a time span that is
considerably shorter than the average career, making life long employment a rarity and
firm and industry specific skills a more risky investment than before.
A closely related phenomenon is the rapid increase in the knowledge intensity of
production, both in industry and services in the North. The trends away from low skilled
manufacturing in OECD labour markets have been well-documented indeed (Katz and
Autor 1999; Juhn et al. 1993; Gottschalk and Moffit 1994; Goldin and Katz 2001;
Nickell and Bell 1996; and Acemoglu 2002a). Figure 9 shows how relative wages have
been decreasing in the United States for low skilled workers in spite of strong increases
in the average educational level, suggesting massive relative demand shifts away from
low skilled labour.
10
Figure 10: Relative wage and supply US 1967–2000
Source: Card and DiNardo (2002).
A similar demand trend, even if wage changes were often less dramatic, has been
identified for virtually every OECD country (see for example Autor et al. (2005) for
recent evidence on the US and Machin and Van Reenen (1998) or Sanders and ter Weel
(2000) for surveys of the evidence on the OECD).
For example Autor et al. (1998) showed that the use of new technology, particularly
new ICT related technologies, was highly correlated with the drop in unskilled labour
demand. The link between technology and labour demand was further explored
empirically by Berman et al. (1994), Machin and van Reenen (1998), Krusell et al.
(2000) and Doms et al. (1997).
Models to explain the technology-skilled labour–demand nexus can be found in
Acemoglu (2002b) and Sanders (2005). Acemoglu attributes skill biases in technical
change to the market size effect, which claims that technology was designed to be using
skilled labour because the skilled labour supply increased in the 1970s. Sanders on the
other hand claims, in accordance with Schultz (1961) that technology is skill biased by
nature when it is new.
This is the essence of a life cycle model and the evidence seems to support the latter
type of model. Greenwood and Uysal (2004) make the case compellingly. The number
of firms per capita and trademarks granted has exploded over the 1980s in the US.
Kortum and Lerner (1997) show a similar speed-up in patenting and most activity
obviously takes place in the new technology fields such as biotech and software. All this
increased activity indicates high levels of grassroots entrepreneurship and product
innovation, which in Greenwood (1997) but also in life cycle models causes temporary
shifts in the relative demand for skilled labour.
Another strand of literature, e.g., Caroli and van Reenen (2001), Bresnahan et al. (2002)
would focus more on the effects of technology through the organization of work, rather
than on characteristics of the new technologies per se. According to their evidence, the
activities within firms have shifted towards design, sales, marketing, etc. whereas
production itself was gradually automated, outsourced, or even off-shored.
11
Piva et al. (2005) find evidence of a mutually reinforcing effect. In a recent and careful
re-evaluation of the evidence for the USA, Autor et al. (2005) conclude that technology
and organizational change tend to ‘polarize’ the demand for labour.4 It is the
intermediate level workers that are being replaced by technology.
Meanwhile changing trade patterns also affected the North. The trade of goods tripled
between 1985 and 1996 while the trade of services increased by more than three times
over this time period. While some of this increase in the world export rate is attributable
to an increased participation in international trade by countries, which had previously
been excluded, export rates in the leading industrialized countries have also increased
over the past three decades. For example, US exports and imports have increased from
11 per cent of GDP in 1970 to more than 25 per cent by 1999. 5 Exports as a share of
GDP for 49 other developed countries have risen from around 18 per cent in 1982 to
around 25 per cent by 1999.
Evidence, collected by Feenstra and Hanson (1996; 1999; 2004), shows that over the
1980s and 1990s the import of intermediate (manufactured) goods has almost doubled
from 6.5 to 11.6 per cent of total intermediate purchases in the US. More open
economies are likely to show much higher levels and similar increases.
The increase in world trade is also not attributable to the influence of just a few
industries or sectors, but rather systematic across most parts of the economy. The
exposure to foreign competition in manufacturing increased by about one-sixth in the
OECD countries. The exposure to foreign competition increased in every single OECD
country, with the exception of Japan. In addition, it increased in most of the
manufacturing industries.
The way we read the evidence is that globalization combined with technological
change, and in particular the information technology and communication technologies
breakthroughs, had rendered the comparative advantage in low technology and even
traditional moderate technology industries incompatible with high wage levels. At the
same time, the emerging comparative advantage that is compatible with high wage
levels is based on innovative and knowledge-intensive activities. There are many
indicators reflecting the shift in the comparative advantage of the high wage countries
towards an increased importance of such activities. For example, the ICT sector of the
USA has experienced an increase in the annual growth rate from 5 per cent in 1991 to
nearly 20 per cent by 1998. By contrast, the rest of the economy experienced fairly
steady growth at around 3 per cent over this period. Before we turn to the role of the
entrepreneur in this ‘new’ economy we establish fact 5:
Fact 5: The North has experienced de-industrialization and a changing relative
labour demand towards skilled non-production workers
Increased globalization of economic activity seemingly also condemned
entrepreneurship, in the form of new firm start-ups and small firms, to extinction. Or at
4 It is likely that the evidence for skill polarization is less pronounced in the European welfare states,
where a large low skilled service sector is absent and skill levels tend to be higher for the least
educated.
5 ‘Markets Go Global’, The Economist, 20 September, 1997.
12
least there seemed to be only an even more diminished role than in the capital driven
Solowian Industrial Economy. Caves (1982: 53) argued that the additional costs of
globalization that would be incurred by small business ‘constitute an important reason
for expecting that foreign investment will be mainly an activity of large firms’.
Certainly the empirical evidence by Horst (1972) showed that even after controlling for
industry effects, the only factor significantly influencing the propensity to engage in
FDI was firm size. As Chandler (1990: 130) concluded, ‘to compete globally you have
to be big’. Thus, it was particularly startling and a seeming paradox, when scholars first
began to document that what had seemed like the inevitable demise of entrepreneurship
actually began to reverse itself starting in the 1970s.
Loveman and Sengenberger (1991) and Acs and Audretsch (1993) carried out
systematic international studies examining the re-emergence of small firms and
entrepreneurship in North America and Europe. Two major findings emerged from
these studies. First, the relative role of small firms varies systematically across
countries, and secondly, in most European countries and in North America, small firms
began increasing their relative importance starting in the mid-1970s. In the USA the
average real GDP per firm increased by nearly two-thirds between 1947 and 1980, from
US$150,000 to US$245,000, reflecting a trend towards larger enterprises and a
decreasing importance of small firms. However, within the subsequent seven years, by
1987, it had fallen by about 14 per cent to US$210,000 (Brock and Evans 1989).
Similarly, small firms accounted for one-fifth of manufacturing sales in the USA in
1976, but by 1986 the small firm share of sales had risen to over one-quarter (Acs and
Audretsch 1990).
It was in the area of job generation that the recent emergence of entrepreneurship was
first identified. In 1981 David Birch revealed the startling findings from his long term
study of US job generation. Birch (1981: 8) found that, ‘Whatever else they are doing,
large firms are no longer the major providers of new jobs for Americans’. Instead, he
discovered that most new jobs emanated from small firms. More recently, Davis et al.
(1996a; 1996b) corrected for the regression to the mean fallacy which they claim is
inherent in Birch’s results. While their quantitative results differ from Birch’s, their
study still indicates that small firms account for more than their share of new
employment. In particular, in their study large firms created 53 per cent of the new jobs
but their employment share is 65 per cent. At the same time, large firms destroyed 56
per cent of the jobs, which is greater than their share of new jobs created. Their measure
was static in nature and gave no indication whether this share has been increasing or
decreasing over time. Still the results indicate that small firms are net job generators,
while large firms shed labour.
The rise in entrepreneurial activity is not particular to the US. The changes in
employment shares in Europe are telling. The small firm employment share in
manufacturing in the Netherlands increased from 68.3 per cent in 1978 to 71.8 per cent
in 1986; in the United Kingdom from 30.1 per cent in 1979 to 39.9 per cent by 1986; in
(West) Germany from 54.8 per cent in 1970 to 57.9 per cent by 1987; in Portugal from
68.3 per cent in 1982 to 71.8 per cent in 1986; in the North of Italy from 44.3 per cent in
1981 to 55.2 per cent by 1987, and in the South of Italy from 61.4 per cent in 1981 to
68.4 per cent by 1987 (Acs and Audretsch 1993). An EIM study documents how the
relative importance of SMEs in Europe (19 countries), measured in terms of
employment shares has continued to increase between 1988 and 2001 (EIM 2002).
13
Methodologies similar to Birch’s were also used in the European context to assess the
contribution to new job generation. In one of the first studies Gallagher and Stewart
(1986) and Storey and Johnson (1987) found similar results for the United Kingdom.
Konings (1995) linked gross job flows in the United Kingdom to establishment size and
finds that the gross job creation rate is the highest in small establishments and the
lowest in large establishments. By contrast, the gross job destruction rate is the lowest in
small establishments and the highest in large establishments.
Evidence from Sweden (Heshmati 2001) also suggests that employment creation is
negatively related to firm size based on data from the 1990s. Similarly, Hohti (2000)
finds that gross employment creation and destruction are negatively related to firm size
in Finland. Using data from Finnish manufacturing between 1980–94, Hohti (2000)
finds that the annual job flow rates, in terms of births and deaths, is similar to that
identified by Broesma and Gautier (1997: 216) for Dutch manufacturing firms and by
Klette and Mathiassen (1996) for Norwegian manufacturing firms. In particular, new
establishments have the greatest job creation rates as well as the greatest rates of job
destruction. Thus, the evidence from Finland, as well as from Sweden, and the
Netherlands, suggests entrepreneurial dynamics similar to those found in North
America.
Hence, the weight of the empirical evidence on employment generation is remarkably
robust and indicates that the role of entrepreneurship in employment generation in
Europe is consistent with the findings for the United States. The reversal of the trend
towards large enterprises and the re-emergence of small firms were not limited to
employment only. For example, in the Netherlands the business ownership rate fell
during the post-war period, until it reached a trough of 0.085 in 1982. But this
downward trend was subsequently reversed, rising to a business ownership rate of 0.10
by 1998 (Audretsch et al. 2002).
Small and new enterprises are therefore increasingly important qualitatively, as a mode
of organization and operation, and quantitatively as an engine of employment creation
on both sides of the Atlantic. In addition it should be emphasized that an important
qualification of the ‘Job Generation’ literature, is that it only links employment changes
of the firm to the size and in some cases the age of the firm. This means that the
performance criterion is not focused on aggregate employment changes, but
employment changes at the level of the firm. This implicitly assumes that there is no
externality or spillover from one enterprise to other firms. At this point we feel
confident to state fact 6:
Fact 6: Entrepreneurial activity has risen
And with this fact we complete our list of facts to explain and be explained in the model
we develop in the next section. Section 2.4 wraps up this section first.
2.4 Summary
This section has established the facts we intend to explain and those we intend to use as
explanation. We intend to explain how political liberalization and technology have
interacted and offered the Northern entrepreneurs the opportunities they needed to
transform trade and production patterns and shape what we know as globalization and
the knowledge economy. We have shown that:
14
1. Labour abundant countries have joined the global economic system
2. Political risk has declined in the world
3. Information and Communication Technologies is a General Purpose Technology
that fuels globalization.
4. Arguably, these developments are exogenous to the economic system we
consider. Starting from these observations we speculated that they may cause:
5. FDI driven industrialization in the South(East)
6. De-industrialization and shifting relative labour demand in the North
7. Increased entrepreneurial activity in innovation, outsourcing, and off-shoring.
The next section presents a model where the link between cause and effect is
formalized.
3 The
model
Aspects involving the decision to locate production in foreign countries are certainly not
a new subject and have been studied in the context of so called product life cycle
models since Vernon coined the term in 1966. He linked the location of production
activities to the life cycle stage of the product, hypothesizing that new products, because
of their higher knowledge intensity, would be produced in the North, that has a relative
abundance of skilled labour and therefore a comparative advantage in their production.
Krugman (1979) formalized Vernon’s idea in a model but assumed that technology
generation and transfer are both exogenous processes. Grossman and Helpman (1991)
used the insights of endogenous growth theories to endogenize the process of innovation
and imitation. In these models, however, no distinction is made between skilled and
unskilled labour. Moreover, Grossman and Helpman are also not explicit about the role
of Northern agents in initiating the transfer of technology across geographic space.
Instead they assume that R&D undertaken in the South drives technology diffusion by
copying ideas off the shelve from the North.
In this tradition we now present a model that distinguishes among three stages in the life
cycle of products. The first stage is initiated by the discovery of a new product. Upon
discovery, an entrepreneur in the North will introduce the new product or service into
the Northern economy. The second stage occurs when the product matures, leading the
incumbent firm to apply firm-based R&D to the standardization and routinization of the
production process, driven by the desire to cut costs. As the production process becomes
standardized, less skilled workers in the North can then be engaged in the production
process at relatively low wages. However, so can Southern workers. Still, we assume
that the act of moving production abroad marks the beginning of a separate third and
final stage in the product life cycle. Setting up production facilities in the South is
neither costless nor riskless and requires an entrepreneurial act. We assume that a
Northern entrepreneur initiates this process as copying behaviour is important especially
after Northern entrepreneurs have set up shops in the South. We therefore assume that it
is FDI, not Southern imitation R&D that will be the medium of knowledge spillover.
15
By endogenizing the allocation of entrepreneurial talent over product innovation,
process innovation, and outsourcing/off-shoring, we can explain the dynamics in global
specialization patterns as the result of political risks falling and low skilled labour
abundant countries joining the global economy. We also show that the introduction of a
so called GPT drives the Northern economy further towards what we have coined the
entrepreneurial economy above; a production structure that relies heavily on new goods
production and product innovation for creating value added.
The model structure follows that of a standard endogenous growth with variety
expansion in a final goods model. First we assume identical time separable preferences
globally and derive consumer demand for goods over time solving:
∞ −ρ(τ−t
max : e
) log(E / P ) τ
∫
τ
τ d
Et
(1)
t
s t
. . : Y + rA = E + A
&
t
t
t
t
where E is expenditure on consumption, P is the price of a unit of direct utility, ρ is the
discount rate, Y is income, A is the level of assets and a dot over a variable signifies the
time derivative. This problem yields the familiar Ramsey result:
E& / E = r − ρ
(2)
t
t
Then we assume a standard Constant Elasticity of Substitution (CES) love-of-variety
instant utility function and solve:
1/ α
n
⎛
⎞
α
max :U =
c di
∫ i
⎜
⎜
⎟
⎟
ci
⎝ 0
⎠ where 0≤ α≤1
(3)
n
s t
. . : c p di ≤ E
∫ i i
0
i indexes a single good and n is the total number of goods consumed. c and p are
quantity and price of a single good in the basket. To derive the instant global demand
functions for all current and future goods in this CES-utility function is
straightforward:6
−
1
α 1
n
1
−
⎛
⎞
⎛
α
α
⎞
D
p
α
E
c
i
=
≡
i
⎜
⎟
where P
p α 1
− di
∫
(4)
⎝
⎜
⎜
⎟
⎟
P ⎠
P
i
⎝ 0
⎠
P is defined as the minimum cost of one unit of direct utility, U.
Now consider production. Following the life cycle literature we assume that products in
their early stages can only be produced in the North. The reason is that early stage
production requires high flexibility and creativity as well as frequent feedback from the
consumer. Northern skilled workers and sophisticated consumers provide the required
6 See for example Grossman and Helpman (1991).
16
setting. All these early stage aspects give skilled workers a competitive advantage in
production (Schultz (1961). As there is also a large tacit knowledge spillover required
from inventor to innovator and early stage producer, it is reasonable to assume that a
new product is initially produced by only skilled workers and only in the North.
At that stage the producer therefore faces no competition, at home or abroad, allowing
him to set monopoly prices:
N
N
max : π = c p − w H l Hi
i
i
i
pi
D
s t
. . : c = c i
(5)
i
N
N
s t
. . : y = b H l Hi
i
where wNH is the high skilled wage in the North, l is the amount of labour employed.
Note also that labour is the only factor of production and the production volume, y is
assumed linear in labour employed. This yields:
N
w H
p =
N
n
∀ ∈ n
(6)
i
N
α
H
b H
As a product matures the incumbent producer will invest in R&D to come up with a
process that allows the product to be produced by low skilled workers in the North.7
This will reduce costs and hence increases profits. At the mature stage, however, the
market will no longer be uncontested. The fired skilled workers can always set up a new
firm and set their price equal to marginal costs:8
N
w H
p =
(7)
i
N
b L
the incumbent, however, using the mature production technology, has lower marginal
costs and can retain the entire market and prevent entry by setting his price equal to:
⎡ wN L wN H
⎤
p = min
,
(8)
i
⎢
− ε
N
N
⎥
⎣ b
α
L
b H
⎦
where ε is close to 0. However, the new production process also creates the possibility
of outsourcing the production. To enable Northern low skilled workers to compete
against the high skilled workers, production must be routinized and many routines are
embedded into machines, interfaces, and procedures. Such codifications, however, can
also be transferred (at some cost) abroad. The Northern producer must therefore now
fear competition from abroad. His only edge is the potentially higher marginal
7 Note that in fact we assume that products only mature as the incumbent producer engages in such
R&D. It is not an automatic process as for example in Krugman (1979) or Arrow (1962).
8 Here we have to assume they cannot as long as the firm employs them. Some non-disclosure
agreements typically ensure that they do not infringe upon the incumbents profits while and even
some time after being employed there.
17
productivity of Northern low skilled labour.9 If profitable entry from abroad is feasible
the incumbent is not likely to engage in that activity. It typically takes an entrepreneur
to set it up, take the risks and, if successful, reap the benefits. Products that have been
outsourced have a price:
⎡ wS L wN L
⎤
p = min
,
i
⎢
− ε
S
N
⎥
⎣αb L b L
⎦
(9)
and in our model compete out production in the North entirely. For given n, nNH, nNL and
nSL the labour markets for Northern skilled, unskilled and Southern labour can now be
assumed to clear, such that the model equilibrates. We require:
n
1
α
α
1
1
α
α
1
N
L H * =
−
1 α
−
1 α
1−
∫α b EP ( N
α w H )
⎛
α−
N
N
1 di = n H ⎜ 1−α
−
1 α
1−α
α
b
EP
(w H ) ⎞
α−1 ⎟
⎜
⎟
n N +n S
L
L
⎝
⎠
1
H
⎛
1
⎞
n −n L
α ⎛
α
N
N
α 1
N
N
α 1
w
w
w
w
N
−
⎡
H
L ⎤ ⎞ −
⎜
⎛
1
H
L
1−
N
−
⎡
⎤ ⎞ − ⎟
L L * =
α
b EP
min
,
di =
∫
⎢
⎥
n L ⎜ 1
1−
⎜
⎜
⎟
⎟
α
b EP
min⎢
,
⎥
⎟
⎜
⎜
⎟
⎟
S
b
b
α
b
αb
n L
⎝
⎣
⎦ ⎠
⎜
⎝
⎣
⎦ ⎠
⎟
⎝
⎠
1
S
⎛
1
⎞
n L
α ⎛
⎡ N
S
w L
w L ⎤ ⎞ α−
α
1
⎜
⎛
⎡ N
S
w L
w L ⎤ ⎞ α−1 ⎟
S
L L * =
−1
1−α
b EP
⎜min
,
⎟
di =
∫
⎢
⎥
S
n L ⎜ −1
1−α
b EP
⎜ min
,
⎟
⎜
⎟
⎢
⎥
⎟
⎜
⎟
b
αb
b
αb
0
⎝
⎣
⎦ ⎠
⎜
⎝
⎣
⎦ ⎠
⎟
⎝
⎠
(10)
Although this set of equations cannot be solved analytically, due to the discontinuities in
the demand curves, one can analyse the equilibrium graphically and sequentially. First
consider the market for high skilled workers in the North in the left panel.
9 In reality there is all kinds of competitive advantages. The Northern incumbent may enjoy lower
transportation costs to the largest markets, protective tariffs and subsidies, complementary sectors, etc.
etc.
18
Figure 11: Labour market equilibrium
Source: compiled by the authors.
There is no competition in this market, implying that prices are always set to profit
maximization levels and consequently the demand curve is continuous. Confronted with
exogenous supply this yields an equilibrium wage at wNH*. Given that wage the
marginal costs in the new goods sector are given by wNH*/b, which puts an upper limit
on prices in the mature Northern sector. Beyond that price demand for goods falls to 0,
so demand for labour falls to 0. Therefore the demand curve has a horizontal section at
the low skilled wage that would imply a profit maximizing price is set equal to wNH*/b.
This is the case at wNL= αwNH*. Then the kinked curve is confronted with exogenous
supply and we obtain the Northern equilibrium low skilled wage. By similar argument
one can also construct the demand for Southern low skilled labour and solve the final
market. All labour markets equilibrate simultaneously for all given combinations of n,
nNH, nNL and nSL. n, nNH, nNL and nSL, however, will not be constant over time. The
assumption of imperfect competition implies there are rents in the model and therefore
there is an incentive for new entrants. Consider the effects on profit flows and
ultimately the value of being a producer in the three stages of the life cycle. With prices
set as above, there are four scenarios that we need to consider. The cost gaps between
high and low skilled labour using producers in the North and between Northern and
Southern low skilled labour using producers can be wide or narrow respectively. Hence
profits for the three types of activity under these conditions are given in Table 1. The
most important result in Table 1 is the fact that profits are always positive and in steady
state equilibrium, where relative wages and relative sector sizes are constant,
proportional to total expenditures and inversely proportional to their own sector size.
Given these profit flows it is therefore worthwhile to engage in entrepreneurial activity,
product development, and outsourcing/off-shoring.10
10 Obviously it is worthwhile only if the discounted present value of this profit flow exceeds the costs of
making the required innovation.
19
Table 1
Profits under various scenarios
Gap
Gap nNH-nNL
nNL-
Wide Narrow
nSL
−1
−
⎛
1
α
α
α
α ⎞
⎛
α
α
α
⎞
− α E ⎜
N
N
−
N
S
S
S
1
N
α
1−
S
S
α
−1
S
α
1
1
(
)
−
n
− α E ⎜
α −1
1−
n L
n L ⎛ w L ⎞
⎛ b L ⎞ α ⎟
L ⎛ w
L ⎞
⎛ b L ⎞
n L ⎛ w L ⎞
⎛ b L ⎞ α ⎟
1
(
)
N
π
N
α −1
π H =
+
α
+
H
=
⎜1+
+
1
N
⎜
N
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟ ⎟
N
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟ ⎟
n H
⎜
n H ⎝ w H ⎠
⎝ b H ⎠
n H ⎝ w H ⎠
⎝ b H ⎠
⎟
n H
⎜
n H
n H ⎝ w H ⎠
⎝ b H ⎠
⎟
⎝
⎠
⎝
⎠
−1
⎛
α
α
α
α
⎞
−1
⎛
α
α
N
α
⎞
− α E ⎜
N
N
−1
N
α
1 −
S
S
α
−1
S
α
1
1
(
)
−
n
− α E ⎜
S
S
−
S
α
−
n H
n L ⎛ w L ⎞
⎛ b
α L ⎞ α ⎟
H ⎛ w
H ⎞
⎛ b H ⎞
n L ⎛ w L ⎞
⎛ b L ⎞ α ⎟
1
1
1
(
)
Wide
N
π
N
1− α
L =
⎜1+
+
⎟
π L =
α
N
⎜1+
+
N
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟ ⎟
N
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟
n L
⎜
n L ⎝ w L ⎠
⎝ b L ⎠
n L ⎝ w L ⎠
⎝ b L ⎠
⎟
n L
n L
n L ⎝ w H ⎠
⎝ b H ⎠
⎝
⎠
⎜
⎟
⎝
⎠
−1
⎛
α
α
α
α
⎞
−1
α
α
α
α
−
⎛
⎞
α E ⎜
N
N
−1
N
α
1−
N
N
α
−1
N
α
1
1
(
)
n
N
N
N
N
N
N
L ⎛ w
L ⎞
⎛ b L ⎞
n H ⎛ w H ⎞
⎛ b H ⎞ −α ⎟
− α E ⎜
α −1
1−α
α −1
1−
n L ⎛ w H ⎞
⎛ b H ⎞
n H ⎛ w H ⎞
⎛ b H ⎞ α ⎟
S
π
1
(
)
S
L =
⎜1+
+
π L =
S
⎜1+
+
S
⎜
⎜ S ⎟⎟ ⎜⎜ S ⎟⎟
S
⎜
⎜ S ⎟⎟ ⎜⎜ S ⎟⎟ ⎟
S
S
⎜
⎜ S ⎟⎟ ⎜⎜ S ⎟⎟
S
⎜
⎜ S ⎟⎟ ⎜⎜ S ⎟⎟ ⎟
n L
⎜
n L ⎝ w L ⎠
⎝ b L ⎠
n L ⎝ w L ⎠
⎝ b L ⎠
⎟
n
n
w
α
⎝
⎠
L
⎜
L ⎝
L ⎠
⎝ b L ⎠
n L ⎝ w L ⎠
⎝ b L ⎠
⎟
⎝
⎠
−1
⎛
−
α
α
α
α ⎞
1
⎛
α
α
N
α
⎞
− α E ⎜
N
N
−1
N
α
1−
S
N
α
−1
S
α
1
1
(
)
−
n
− α E ⎜
S
N
−1
N
α
1−
n L
n L ⎛ w L ⎞
⎛ b L ⎞ α ⎟
L ⎛ w
L ⎞
⎛ b L ⎞
n L ⎛ w L ⎞
⎛ b L ⎞ α ⎟
1
(
)
N
π
N
α −1
π H =
+
α
+
H =
1
N
⎜1
N
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟ ⎟
N
⎜ +
+
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟ ⎟
n H
⎜
n H ⎝ w H ⎠
⎝ b H ⎠
n H ⎝ w H ⎠
⎝ αb H ⎠
⎟
n H
⎜
n H
n H ⎝ w H ⎠
⎝ αb H ⎠
⎟
⎝
⎠
⎝
⎠
−1
⎛
α
α
−1
⎛
α
α
N
α
⎞
N
N
N
S
α ⎞
− α E ⎜
α −1
1
1
(
)
−
n
− α E ⎜
S
N
−1
N
α
1−
n H
n L ⎛ w L ⎞
⎛ b L ⎞ α ⎟
H ⎛ w
H ⎞
⎛ b H ⎞ α n L
⎟
1
(
)
Narrow
N
1
π
N
1−
π L =
+
α
α
+
L =
α
1
N
⎜
N
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟ ⎟
N
⎜1
−
+
+
N
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟
α
N
⎟
n L
⎜
n L ⎝ w L ⎠
⎝ b L ⎠
n L
⎟
n L
⎜
n L
n L ⎝ w H ⎠
⎝ b H ⎠
⎟
⎝
⎠
⎝
⎠
−1
⎛
α
α
−1
⎛
α
α
α
α ⎞
α
⎞
− α E ⎜
N
N
N
−1
N
α
1
1
(
)
−
n
− α E ⎜
N
N
−1
N
α
1−
N
N
α
−1
N
α
1−
n L ⎛ w H ⎞
⎛ b H ⎞
n H ⎛ w H ⎞
⎛ αb H ⎞ α ⎟
L
n H ⎛ w H ⎞
⎛ αb H ⎞ α ⎟
1
(
)
S
1
π
S
π L =
+
+
L =
α
1
S
⎜
S
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟
S
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟ ⎟
S
⎜1
−
+
α +
S
S
⎜
⎜ N ⎟⎟ ⎜⎜ N ⎟⎟ ⎟
n L
⎜
n L
n L ⎝ w L ⎠
⎝ b L ⎠
⎟
n L
⎜
n L ⎝ w L ⎠
⎝ b L ⎠
n L ⎝ w L ⎠
⎝ b L ⎠
⎟
⎝
⎠
⎝
⎠
20
Entrepreneurial activity introduces new goods and services into the economy. Product
development allows existing products to be produced by low skilled labour in the North
and outsourcing moves the production of routinized products to the South. Even if
patents are perfectly enforced, such activities are worthwhile, as the flow of profit
increases with each stage in the life cycle, as long as wNH > wNL > wSL. The incentive to
move the product over the cycle then comes from increments in the profit flow. This
also explains why incumbents may engage in (defensive) product development and
outsourcing even though they cannibalize on their current profits.11 The incentive to
become an innovating entrepreneur is the value of being a Northern uncontested new
good producer that is given by:
T
N
−r (τ t )
v
=
H
E
e
π N H (τ) τ
d I (t)
∫ −
(11)
t
the expected discounted profit flow from time t, the time of entry, to time T, the time at
which production with low skilled labour starts and monopoly profits are contested,
conditional on the information set I(t) available at time t. The flow probability of being
contested is given by the relative gross number of product development,
( N
S
n L
&
+ n& ) N
L
/ n H . Expecting that profit grows at a constant rate, as it will in the steady
state, yields:
∞
N
N
S
N
N
N
−( +( & + & ) /
)( − )
( & / − &
/
)( − )
π
(t)
N
N
r
n L n L
n H
τ t
N
E E n H n
τ
H
t
v
H
=
=
H
π H (t) e
π H (t)e
τ
d
∫
N
r − E& / E + n& / n H
t
(12)
An entrepreneur is able to capture this value by setting up a firm and starting production
of a new good or service. Similarly we have for the Northern low skilled producers:
N
π L (t)
N
v
=
L
(13)
S
N
N
r − E& / E + (n
+
L
&
n L
& ) / n L
where the current profits are discounted by the rate of interest, r, the rate of profit
erosion, − (
N
N
E& / E − n L
& / n L ) and the risk of losing the profits because of outsourcing,
S
N
n L
& / n L . However, the value of a product development to the incumbent is lower. In
order to obtain the higher profit flow an incumbent foregoes the existing profit flow.
Hence the value of a product development equals the positive difference between (13)
and (12). Finally the value of owning an outsourced firm that produces in the South is
equal to:
S
π L t
( )
S
v
=
L
(14)
S
S
r + ψ − E& / E + n L
& / n L
where ψ is a risk premium for producing abroad that reflects the risks of being unable to
retrieve profits and reflects the perhaps less than perfect property rights protection in the
11 In fact we assume their cost advantage and tacit knowledge give them a sufficient edge, so we see
only incumbent firms doing product development.
21
South.12 Once more the value to the incumbent and therefore also to any entrant that is
forced to compensate him for patent infringement, equals the positive difference
between (14) and (13).13
Let us finally assume that introducing a new good or service, developing a product and
outsourcing it all draw on the same pool of limited resources in the North, which we
may label entrepreneurial talent, T.14 To close the model we must specify how
entrepreneurial talent combines with accumulated knowledge to generate the dynamics
in the model. Assume:
N
N
S
γ
A
N
S
n&
= & − & − & =
− & −
H
n
n L
n L
aT n
n L
n L
&
I
N
γ
N
S
n&
=
−
L
aT n H
n L
&
where 0≤γ≤1
(15)
D
S
γ
N
n&
=
L
aT n L
O
Note that a productivity parameter a and the amount of entrepreneurial talent allocated
are combined with nA, nNH and nNL to reflect the knowledge spillovers in the model.
Innovation benefits from the (exogenous) stock of knowledge, nA.15 Development
benefits from the knowledge accumulated in new goods production, nNH, and
outsourcing benefits from experience in routinized production, nNL. Furthermore (15)
reflects the assumption that a product must first be invented, can then be routinized and
finally outsourced, but only in that order, missing none of the stages in its development.
Entrepreneurial talent now has to be allocated over three alternative uses. In each it
must generate the same marginal value product. Assuming this (market clearing) reward
to entrepreneurial talent is given by wT we obtain:
γ−1 A N
γ−1 N
−
w = γaT
n v
=
−
=
1
−
H
γaT
n H v
v
γaT
n
v
v
T
I
D
( N
N
L
H )
γ
N L
O
( S
N
L
L )
(16)
This arbitrage condition can be rewritten in separate entrepreneurial talent demand
equations that are downward sloping in the reward for entrepreneurial talent.16 Equating
12 This is our parameter for political risk.
13 Results do not change qualitatively and become stronger if we allow for entrepreneurs to not
compensate incumbents. Obviously that makes outsourcing even more attractive.
14 Alternatively one could think of R&D labour, engineering talent, or organizational and managerial
talents. All that is required to justify this assumption is that all innovative activities require the same
labour that is a different type of labour than that used in production.
15 See Sanders (2005) for more details on how this knowledge parameter can be endogenized. In
standard new growth models it is frequently assumed to be equal to n, accumulated innovation. In this
model this would also introduce endogenous growth but prevent us from experimenting with shocks to
the knowledge stock. The introduction of a new GPT, however, is exactly that, a positive shock to nA.
16 One could consider setting wT equal to the high skilled wage in the North as it arguably presents the
opportunity costs for the entrepreneur. However, this complicates the model considerably without
adding much in terms of insight in the mechanisms we are primarily interested in. Sanders (2005)
shows the equivalence in qualitative results in a different (two stage life cycle) but closely related
model.
22
the sum to the total stock of entrepreneurial talent in the North yields the equilibrium
reward for entrepreneurial activity:
γ 1
( −γ)
1
1
⎛
⎞
N
N
1−γ
N
S
N
1−γ
⎜
⎛
⎛
⎞⎞
⎛
⎛
⎞⎞ ⎟
γ−
n
v
n
v
v
1
H
L
L
L
L
w = aT *
A
N
n v
+
−
+
−
H ⎜1
1
⎜
⎜
⎜
⎜
⎟
⎟⎟⎟
⎜
⎜
⎟
⎟
⎟
T
A
N
⎜
⎜ A
N
N
⎟
⎟
n
v H
n
v H
v H
⎜
⎝
⎝
⎠⎠
⎝
⎝
⎠⎠ ⎟
⎝
⎠
(17)
which in steady state grows in proportion to nA times vNH. As the latter grows in
proportion to profits, which in turn grow in proportion to expenditure, E, over nNH in the
steady state the reward to entrepreneurial talent will grow in proportion to total
expenditure, as will wages in all labour markets, as long as nA grows at the same rate as
nNH. Normalization of expenditure then pins down the growth rate of all nominal
variables in the model to 0. The total range of goods expands at the common growth
rate of new, mature, and outsourced goods in the global economy. This leads to
propositions I and II.
Proposition I: There exists a steady state equilibrium in which the allocation of all
labour types is stable, all ranges of goods grow at the common positive rate and labour
markets are clear.
Proposition II: This steady state is a unique and globally stable equilibrium in the sense
that the economy will converge back to this steady state when it is disturbed by shocks.
The proof for both propositions is in the Appendix. The dynamic equilibrium in the
model can be illustrated by plotting the three demand equations for entrepreneurial
activity, implicit in equation (16) and equating the vertical sum of these curves to the
exogenous supply of entrepreneurial talent in the North. In the steady state these levels
of activity must generate the same rate of expansion G*, to the respective variety
ranges. The knowledge spillover structure assumed in equation (15), which implies all
entrepreneurial activity receives a spillover from upstream entrepreneurial activity
(which for innovation is exogenous knowledge creation), guarantees that in the end all
variety ranges will grow at the same rate for a stable allocation of entrepreneurial talent.
As was shown under Proposition II and in the Appendix this implies in the steady state
the right panel of Figure 11 applies.
23
Figure 12: Steady state equilibrium
Now consider the comparative statics in the steady state concerning the effects of
increases in LSL, (globalization/integration), ψ (property rights/institutions) and nA
(technology shock/GPT) and our policy experiment to encourage and stimulate
entrepreneurship T* (education/entrepreneurship). To see what happens consider first
the shocks in isolation.
A sufficiently large increase in southern labour supply (if the shock is too small one
may stay in a situation of narrow wage gap and relative profits do not change, causing
no response) will initially drive down equilibrium wages in the South and push the
southern economy out of a possible narrow gap situation. In Table 1 it can be seen that
in that case profits in the South will rise as wages fall. By equation (14) that implies the
value of outsourcing activities increases and entrepreneurs will switch out of innovation
and development to outsource more mature products to the South. As the rate of
outsourcing accelerates and the rates of innovation and development fall, there is a
change in the relative composition of the economy. More varieties are produced in the
South, variety expansion falls and both nNH and nNL will fall relative to nSL. This, by (10)
implies that Southern wages recover and Northern wages will fall, whereas the reward
to entrepreneurial talent rises. Of course equilibrium is re-established once diminishing
returns in innovation, development and outsourcing (through the gradual reduction of
upstream knowledge stocks) make sure all ranges grow at the same rate again. It should
be noted here that the prediction that Southern wages fall and recover gradually does not
apply for the countries that join world markets. They enter the market with wages below
wSL and see an immediate rise followed by a further increase in the wage level. The
predictions for Southern wage levels therefore rather apply to for example countries like
Brazil, Mexico, and Egypt than India or China.
The impact of a drop in political risk ψ enters our model directly in the return to
outsourcing equation (14). Decreasing risk implies lower discount rates of given profits,
making outsourcing more attractive to the entrepreneurs. The impact on the allocation
of entrepreneurial talent and the relative composition of the economy is similar as
above. However, there is no steady state reduction in Southern wages. Instead the wages
rise due to the increased demand. In the North the implications for relative wages and
24
income are similar as before. The entrepreneurs gain and the low skilled lose out most
but also high skilled wages fall relative to Southern wages.
A rise in the knowledge stock nA or the rate of knowledge expansion has a temporary or
permanent effect on the steady state composition of the economy, respectively. If there
is a level shock the benefits will dissipate gradually. The high skilled workers in the
North benefit first. Entrepreneurs move into innovation and abandon development and
outsourcing. As the available stock of new products expands, however, the
entrepreneurs will return to development and ultimately outsourcing. The steady state
has no changes so the economy will eventually return to its initial equilibrium. If the
growth rate has increased permanently that implies that in the steady state all ranges of
goods will grow at a higher rate, but also that the steady state size of the new product
range will be larger. Consequently the high skilled in the North gain relative to the low
skilled in North and South and again the entrepreneurs benefit.
A rise in T* the stock of entrepreneurial talent/resources in the Northern economy, has a
peculiar steady state impact in this model. The entrepreneurs will be allocated to
expanding the three variety ranges and these ranges and all ranges will therefore expand
faster for some time, but as the new and mature goods range are also eroded faster only
the range of Southern varieties will grow in the steady state. That generates the
interesting result that the benefits of entrepreneurship training in general leak to
developing countries. As entrepreneurs pass the buck, the buck moves faster to where it
stops. However, one should realize that in passing the buck the entrepreneur
appropriates part of the surplus with every pass. Entrepreneurial output thus increases
and even though their wages fall, their incomes rise to the extent that Northern income
increases relative to Southern income if demand elasticities are sufficiently low. One
should also realize that the North is composed of many countries and any one Northern
entrepreneur may take a mature good and outsource it. Having more entrepreneurs and
fewer workers in your country puts you in a favorable position relative to the other
Northern countries, not in the least because more entrepreneurs also means quicker
adjustment to external shocks as the ones described above. From a global perspective
more entrepreneurs are also unambiguously beneficial. It will reduce global wage if not
income inequality and ensures faster adjustment to equilibrium when exogenous shocks
hit the economy. As the world may still have to deal with such shocks in the future, the
end of oil reserves, nuclear disaster in the Middle East, the recovery of Japan or the
introduction of cold hydrogen fusion reactors. Whatever lays ahead, entrepreneurs can
help deal with it and adjust as long as change opens up opportunities.
Finally consider the effect of all shocks described above combined. As we argued in
Section 1 that is precisely what happened over the past 25 years. The model then
predicts significant wage divergence in the North between the skilled and unskilled and
certainly between the entrepreneurs and workers in general, increasing FDI and
outsourcing to the South, and a widening of the North-South wage gap. The latter has
not been observed in the data as China and India have always been considered part of
the South but taking them out of the sample one sees significant divergence between
OECD and non-OECD countries in both wage and income levels. China and India, on
the other hand, are seen to quickly close the wage and income gap with the North and
not surprisingly they receive the bulk of FDI outflow from the North as well. Lower
political risk implies that the downward wage pressure in the (rest of) the South is
somewhat ameliorated. Our policy experiment of increasing the entrepreneurial
capabilities shows that self-interested national policies may yet help to tip the scales and
25
help close the North-South wage and income gaps while offering new perspectives for
countries in the North to remain competitive among their peers in globalizing markets.
4 Conclusions
Globalization and technological change have triggered fundamental changes in both the
developed and developing countries. This paper illustrates one mechanism through
which these shocks interact. By opening up new opportunities for entrepreneurs new
technology and a global market both create a dynamic that shifts the specialization
pattern in the global economy. An important implication of our model is that developing
countries are gaining the comparative advantage for economic activity in the latter
stages of the product life cycle, where the classic factors of raw labour and physical
capital play an important role. At the same time, globalization has facilitated a shift in
comparative advantage in the advanced economies away from mature industries and
towards economic activity in the early stage of the product life cycle. This reorientation
towards early stage activity was reinforced by the simultaneous rise of a GPT that offers
both the means for organizing off-shoring and outsourcing and presented entrepreneurs
with many opportunities for new products and services. The coincidence of these trends
caused and forced the advanced countries to enter the entrepreneurial economy, whereas
it allowed (some) developing countries to rapidly industrialize. Whether the emergence
of the entrepreneurial economy is permanent or transitory depends upon the nature of
globalization. It does not seem likely, however, based on the model presented in this
paper, to expect a return to the industrial economy any time soon. The political
turnaround we have identified as the exogenous shock driving globalization is not likely
to be reversed. On the technology side, however, we do know that related new products
and services will mature. The plight of the unskilled worker in the advanced economies
may yet be ameliorated by his increased participation in these new industries. Under
current global competitive conditions, however, it is more likely that these industries
will also quickly transfer routine operations to less expensive regions and alternative
employment may be required after all.
The policy implications of our analysis are thus that developed countries’ governments
should gear their labour market and social security policies to the new economic reality.
Flexibility, mobility, and employability are key assets for workers and self-employed in
the entrepreneurial economy. A large and easily accessible low skilled non-tradables
sector such as personal services would have to be developed if welfare states are to
survive this pressure.
For the developing world the focus should gradually shift from attracting FDI inflows to
building local capabilities and absorbing mature technologies. In the end, the transition
to a developed country is possible, but to take-off a strategy of gradually moving
upstream in the life cycles seems appropriate. Modern growth theory has little to offer
for the situation they find themselves in. More R&D will not help regions and countries
in which not even mature routine industrial activity has been developed. But
governments should also realize that comparative and competitive advantages are
temporary in nature and should not be protected against such dynamics. Stimulating
local entrepreneurs to take advantage of the opportunities in global markets is the one
policy that can be shown to deliver lasting advantages as, by their flexibility, these
entrepreneurs can help adapt to possible future shocks as well as the current slump in
global economic growth.
26
Our model does not generate very detailed policy advice for either side but the general
direction is clear and further research requires a fundamental understanding of these
macroprocesses. In that understanding we hope to have contributed.
Appendix
Proof of Proposition I
The proof follows from assuming this steady state exists and showing that it constitutes
an equilibrium in all labour markets. First consider the markets for production labour.
They clear iff the allocation satifies:
1
α
α
N
N
1−α
1
L H *
−α
−
=
α
n H α
b
EP
( N
w H ) 1
1
α 1
−
1
1
α
N
N
α 1
−
α
α
N
α 1
⎛
⎡w H w L ⎤⎞
⎛
⎡w H
⎤⎞
N
N
−
N
1
−
1
1−α
1−α
1
L L * = n Lb EP
min
,
−
=
α
n Lb
EP
min
,
N
⎜
⎜
⎢
⎥⎟⎟
⎢
w
N
⎥
( L) 1α1−
⎜
⎜
⎟
⎟
⎝
⎣ b
αb ⎦⎠
⎝
⎣ w L α⎦⎠
1
1
α
N
S
α 1
−
α
α
N
α 1
⎛
⎡ w L w L ⎤⎞
⎛
⎡w L
⎤⎞
N
S
−
α
S
1
−
1
1−
1−α
1
L H * = n Lb EP
min
,
−
=
α
n Lb
EP
min
,
S
⎜
⎜
⎢
⎥⎟⎟
⎢
w
S
⎥
( L) 1α1−
⎜
⎜
⎟
⎟
⎝
⎣ b
αb ⎦⎠
⎝
⎣ w L α⎦⎠
(A1)
Taking growth rates on both sides implies:
N
N
n&
&
&
H
E
α P
1
w H
&
0 =
+
+
+
N
N
n
1 −
−
H
E
α P
α 1 w H
N
N
n&
&
&
L
E
α P
1
w L
&
0 =
+
+
+
N
N
n
1 −
−
L
E
α P
α 1 w L
(A2)
S
S
n&
&
&
L
E
α P
1
w L
&
0 =
+
+
+
S
S
n
1 −
−
L
E
α P
α 1 w L
S
N
N
w L
&
w L
&
w H
&
=
=
S
N
N
w L
w L
w H
where the latter constraints follows from the fact that relative wages appear in the
expressions. From these conditions we can derive that in any steady state equilibrium
wages must grow at the same rate and hence so must all sectors. This in turn implies in
(15) that the number of firms in each sector must grow at the same constant rate as nA.
Now consider the market for entrepreneurial talent. It clears when (17) holds. Under the
conditions derived above, the reward to entrepreneurship can be shown to grow at rate:
w
&
n A
&
v N
&
&
&
&
&
&
&
H
n A
π N H
n A
E
n N H
E
T
=
+
=
+
=
+
−
=
(A3)
w
n A
v N H
n A
π N H
n A
E
n N H
&
E
T
This implies that, if relative wages are to be constant, all must grow at the same rate as
total expenditures and (A2) reduces to:
27
α P&
α E&
0 = g +
+
(A4)
1 − α P
α −1 E
If we then normalize expenditure to 1 growth rates of profits and consequently
innovation values reduce to –g and the interest rate is endogenously determined by (2)
to equal ρ. There is therefore a steady state equilibrium that is fully characterized by:
n A
&
nN H
&
nN L
&
nS L
&
π N H
π N L
π S L
v N H
&
&
&
&
vN L
&
vS L
=
=
=
= −
= −
= −
= −
= −
= − &
= g
n A
nN H
nN L
nS L
π N H
π N L
π S L
v N H
vN L
vS L
E&
wN H
&
wN L
&
wS L
&
w
&T
=
=
=
=
= r − ρ = 0
E
wN H
wN L
wS L
wT
P&
α −1
=
g
P
α
(A5)
Q.E.D.
Proof of Proposition II
Consider being out of equilibrium such that nNH/nNL>(nNH/nNL)SS. Then by Table 1
πNH/πNL<(πNH/πNL)SS and consequently by (12)–(13) vNH/vNL<(vNH/vNL)SS which implies
that T
SS
SS
N
N
&
/
<
N
N
& /
<
I<TI
and TD>TD such that n H n H
g and n L n L
g . Therefore we
move towards the equilibrium as (15) then implies that nNH/nNL falls towards
(nNH/nNL)SS. That process will end in the steady state. A similar argument can be made
for all possible deviations from the steady state allocation of entrepreneurial activity,
which proves that the steady state is both unique and stable.
Q.E.D.
28
References
Acemoglu, D. (2002a). ‘Technical Change, Inequality and the Labor Market’. Journal
of Economic Literature, 40 (1): 7–72.
—— (2002b). ‘Directed Technical Change’. Review of Economic Studies, 69 (241):
781–809.
Acs, Z., and D. Audretsch (1990). Innovation and Small Firms. Cambridge; MA: MIT
Press.
—— (1993). Small Firms and Entrepreneurship: An East-West Perspective.
Cambridge: Cambridge University Press.
Antras, P. (2005). ‘Property Rights and the International Organization of Production’.
The American Economic Review, 95 (2): 25–32.
Arrow, K. (1962). ‘The Economic Implications of Learning by Doing’. Review of
Economic Studies, 29 (3): 155–73.
Audretsch, D. (2007). The Rise of the Entrepreneurial Economy. Berlin: Springer
Verlag.
Audretsch, D., M. A. Carree, A. J. van Steel, and A. R. Thurik (2002). ‘Impeded
Industrial Restructuring: The Growth Penalty’. Kyklos, 55 (1): 81–98.
Audretsch, D., and Z. Acs (eds) (2005). Handbook of Entrepreneurship Research: An
Interdisciplinary Survey and Introduction. Vol. 1. Berlin: Springer Verlag.
Autor, D. H., L. F. Katz, and M. S. Kearney (2005). ‘Trends in US Wage Inequality:
Re-Assessing the Revisionists’. Working Paper 11627. Cambridge, MA: National
Bureau of Economic Research.
Autor, D. H., L. F. Katz, and A. B. Krueger (1998). ‘Computing Inequality: Have
Computers Changed the Labor Market?’. Quarterly Journal of Economics, 113 (4):
1169–1213.
Berman, E., J. Bound, and Z. Griliches (1994). ‘Changes in the Demand for Skilled
Labor in US Manufacturing Industries: Evidence From The Annual Survey of
Manufactures’. Quarterly Journal of Economics, 109 (2): 367–97.
Birch, D. (1981). ‘Who Creates Jobs?’. The Public Interest, 65 (Fall): 3–14.
Bresnahan, T. F., and M. Trajtenberg (1995). ‘General Purpose Technologies “Engines
of Growth”?’. Journal of Econometrics, 65 (1): 83–108.
Bresnahan, T. F., E. Brynjolfsson, and L. M. Hitt (2002). ‘Information Technology,
Workplace Organization and the Demand for Skilled Labor: Firm-level Evidence’.
Quarterly Journal of Economics, 117 (1): 339–76.
Brock, W. A., and D. S. Evans (1989). ‘Small Business Economics’. Small Business
Economics, 1 (1): 7–20.
Broesma, L., and P. Gautier (1997). ‘Job Creation and Job Destruction by Small Firms:
An Empirical Investigation for the Dutch Manufacturing Sector’. Small Business
Economics, 9 (3): 211–24.
29
Card, D., and J. E. DiNardo (2002). ‘Skill-Biased Technological Change and Rising
Wage Inequality: Some Problems and Puzzles’. Journal of Labor Economics, 20 (4):
733–83.
Caroli, E., and J. Van Reenen (2001). ‘Skill-Biased Organizational Change? Evidence
from a Panel of British and French Establishments’. The Quarterly Journal of
Economics, 116 (4): 1449–92.
Chandler, A. (1990). Scale and Scope: The Dynamics of Industrial Capitalism.
Cambridge, MA: Harvard University Press.
Caves, R. (1982). Multinational Enterprise and Economic Analysis. Cambridge:
Cambridge University Press.
Davis, S. J., J. Haltiwanger, and S. Schuh (1996a). Job Creation and Destruction.
Cambridge, MA: MIT Press.
—— (1996b). ‘Small Business and Job Creation: Dissecting the Myth and Reassessing
the Facts’. Small Business Economics, 8 (4): 297–315.
Doms, M., T. Dunne, and K. R. Troske (1997). ‘Workers, Wages and Technology’. The
Quarterly Journal of Economics, 112 (1): 253–90.
EIM (2002). ‘SMEs in Europe. The European Observatory for SMEs’. Vol. 2. Report
submitted to the European Commission, Enterprise Directorate-General. Düsseldorf:
KPMG Special Services and EIM Business School & Policy Research.
Feenstra, R. C., and G. H. Hanson (1996). ‘Globalization, Outsourcing and Wage
Inequality’. The American Economic Review, 86 (2): 240–45.
—— (1999). ‘The Impact of Outsourcing and High-Technology Capital on Wages:
Estimates for the United States, 1979–1990’. The Quarterly Journal of Economics,
114 (3): 907–40.
—— (2004). ‘Global Production Sharing and Rising Inequality: A Survey of Trade and
Wages. In J. Harrigan, and E. K. Choi (eds), Handbook of International Trade.
London: Basil Blackwell, 146–85.
Florida, R. (2003). The Rise of the Creative Class. New York: The Perseus Book Group.
Gallagher, C., and H. Stewart (1986). ‘Jobs and the Business Life-cycle in the U.K’.
Applied Economics, 18 (8): 875–900.
Goldin, C., and L. F. Katz (2001). ‘Decreasing (and then Increasing) Inequality in
America: A Tale of Two Half-Centuries’. In F. Welch (ed.), The Causes and
Consequences of Increasing Income Inequality. Chicago: University of Chicago
Press.
Gottschalk, P., and R. Moffit (1994). ‘The Growth in Earnings Instability in the U.S.
Labor Market’. Brookings Papers on Economic Activity, 2: 217–72.
Greenwood, J., and G. Uysal (2004). ‘New Goods and the Transition to a New
Economy’. Working Paper 10793. Cambridge, MA: National Bureau of Economic
Research.
30
Greenwood, J. (1997). The Third Industrial Revolution: Technology, Productivity and
Income Inequality. AEI Studies on Understanding Economic Inequality. Washington,
DC: AEI Press.
Grossman, G. M., and E. Helpman (1991). Innovation and Growth in the Global
Economy. Cambridge, MA: MIT Press.
Helpman, E. (1998). General Purpose Technologies and Economic Growth. Cambridge,
MA: MIT Press.
Heshmati, A. (2001). ‘On the Growth of Micro and Small Firms: Evidence from
Sweden’. Small Business Economics, 17 (3): 213–28.
Hohti, S. (2000). ‘Job Flows and Job Quality by Establishment Size in the Finnish
Manufacturing Sector, 1980–1994’. Small Business Economics, 15 (4): 265–81.
Horst, T. (1972). ‘Firm and Industry Determinants of the Decision to Invest Abroad: An
Empirical Study’. Review of Economics and Statistics, 54 (3): 258–66.
Iversen, T., and F. Rosenbluth (2006). ‘The Political Economy of Gender: Explaining
Cross-National Variation in the Gender Division of Labor and the Gender Voting
Gap’. American Journal of Political Science, 50 (1): 1–19.
Juhn, C., K. M. Murphy, and B. Pierce (1993). ‘Wage Inequality and the Rise in
Returns to Skill’. Journal of Political Economy, 101 (3): 410–42.
Katz, L. F., and D. H. Autor (1999). ‘Changes in the Wage Structure and Earnings
Inequality’. In O. Ashenfelter, and D. Card (eds), Handbook of Labor Economics.
Vol. 3. North Holland: Elsevier.
Keller, W. (1996). ‘Absorptive Capacity: On the Creation and Acquisition of
Technology in Development’. Journal of Development Economics, 49: 199–227.
Klette, T. J., and A. Mathiassen (1996). ‘Job Creation, Job Destruction and Plant
Turnover in Norwegian Manufacturing’. Annales d’Economie et de Statistique,
41/42: 97–125.
Konings, J. (1995). ‘Gross Job Flows and the Evolution of Size in U.K.
Establishments’. Small Business Economics, 7 (3): 213–20.
Kortum, S., and J. Lerner (1997). ‘Stronger Protection or Technological Revolution:
What is Behind the Recent Surge in Patenting?’. Working Papers 6204. Cambridge,
MA: National Bureau of Economic Research.
Krugman, P. (1979). ‘A Model of Innovation, Technology Transfer and the World
Distribution of Income’. Journal of Political Economy, 87 (2): 253–66.
Krusell P., L. E. Ohanian, J.-V. Ríos-Rull, and G. L. Violante (2000). ‘Capital-Skill
Complementary and Inequality: A Macroeconomic Analysis’. Econometrica, 68 (5):
1029–53.
Lane, P., and G.-M. Milesi-Ferretti (2003). ‘International Financial Integration’.
Discussion Paper 3769. London: Centre for Economic Policy Research.
Loveman, G., and W. Sengenberger (1991). ‘The Re-Emergence of Small-Scale
Production: an International Comparison’. Small Business Economics, 3 (1): 1–37.
31
Machin, S., and J. van Reenen (1998). ‘Technology and Changes in Skill Structure:
Evidence from Seven OECD Countries’. Journal of Quarterly Economics, 113 (4):
1215–244.
Mayer, J., and A. Wood (1999). ‘South Asia’s Export Structure in a Comparative
Perspective’. Working Paper 91. Sussex: Institute of Development Studies.
Nickell, S., and B. Bell (1996). ‘Changes in the Distribution of Wages and
Unemployment in OECD Countries’. The American Economic Review, 86 (2): 302–
08.
OECD (2005). Micro-Policies for Growth and Productivity: Final Report. Paris:
OECD.
Piva, M., E. Santarelli, and M. Vivarelli (2005). ‘The Skill Bias Effect of Technological
and Organisational Change: Evidence and Policy Implications’. Research Policy, 34
(2): 141–57.
Sanders, M. (2005). Technology and the Decline of Unskilled Labour Demand: A
Theoretical Analysis of the US and European Labour Markets. Cheltenham: Edward
Elgar.
Sanders, M., and B. ter Weel (2000). ‘Skill-Biased Technical Change: Theoretical
Concepts, Empirical Problems and a Survey of the Evidence’. Working Papers 00-8.
Aalborg: Danish Research Unit for Industrial Dynamics.
Schultz, T. W. (1961). ‘Investment in Human Capital’. American Economic Review, 5
(1): 1–17.
Storey, D. J., and S. G. Johnson (1987). Job Generation and Labour Market Change.
London: MacMillan.
Thurow, L. C. (2002). Fortune Favors the Bold: What We Must Do to Build a New and
Lasting Global Prosperity. Cambridge, MA: MIT Press.
Torres, R. (2001). Towards a Socially Sustainable World Economy: An Analysis of the
Social Pillars of Globalisation. Geneva: International Labour Office.
Vernon, R. (1966). ‘International Investment and International Trade in the Product
Cycle’. Quarterly Journal of Economics, 80 (2): 190–207.
UNCTAD (2003). World Investment Report. New York and Geneva: United Nations
Conference on Trade and Development.
WTO (2005). International Trade Statistics. Geneva: World Trade Organization.
32
