Professional Documents
Culture Documents
Regional science analyses important issues surrounding the growth and development of
urban and regional systems and is emerging as a major social science discipline. This series
provides an invaluable forum for the publication of high-quality scholarly work on urban and
regional studies, industrial location economics, transport systems, economic geography and
networks.
New Horizons in Regional Science aims to publish the best work by economists, geogra-
phers, urban and regional planners and other researchers from throughout the world. It is
intended to serve a wide readership including academics, students and policymakers.
Titles in the series include:
Edited by
Robert Stimson
Professor of Geographical Sciences and Planning, University
of Queensland, Australia
Roger R. Stough
Vice President for Research and Economic Development,
NOVA Endowed Chair and Professor of Public Policy, George
Mason University, USA
Peter Nijkamp
Professor of Regional, Urban and Environmental Economics,
VU University Amsterdam, the Netherlands
Edward Elgar
Cheltenham, UK Northampton, MA, USA
Published by
Edward Elgar Publishing Limited
The Lypiatts
15 Lansdown Road
Cheltenham
Glos GL50 2JA
UK
Index 325
vii
for Social Science Research, and was the Convenor of the Australian
Research Council Research Network in Spatially Integrated Social
Science (ARCRNSISS) until 2009. He is a Fellow of the Academy
of the Social Sciences in Australia, a past President of the Regional
Science Association International, and the current Chair of the Applied
Geography Commission in the International Geographical Union.
Roger R. Stough is the Vice President for Research and Economic
Development at George Mason University, USA, where he also holds
the NOVA Endowed Chair and is an Eminent Scholar and Professor of
Public Policy. He is a past President of the Regional Science Association
International and holds an honorary doctorate from Jonkoping University
in Sweden.
Michael Taylor is Professor of Human Geography at the University of
Birmingham, UK. He has held academic positions in New Zealand and
Australia, and also worked for the Australian Federal Administration in
Canberra. His research is focused on enterprise and regional economic
development and particularly on the theory of the firm. He is Chair of
the International Geographical Union Commission on the Dynamics of
Economic Spaces.
Tojo Thatchenkery is Professor and Director of the Masters in Science
program in Organization Development and Knowledge Management
at the School of Public Policy, George Mason University, USA. He is a
member of the NTL (National Training Laboratory) Institute for Applied
Behavioral Science, the Taos Institute, the Editorial Board of the Journal
of Applied Behavioral Sciences, the Journal of Organizational Change
Management (JOCM), and the book review editor of JOCM.
Frank Vanclay is Professor of Cultural Geography at the University of
Groningen, the Netherlands. He is a past President of the International
Rural Sociology Association (IRSA).
xi
INTRODUCTION
Over the past few decades the emphasis in regional development theory
has shifted from a focus primarily on exogenous factors to an increasing
focus on endogenous factors. Traditional regional economic development
approaches were erected on neoclassical economic growth theory, based
largely on the Solow (1956, 2000) growth model. The emerging endogen-
ous approaches while recognizing that development is framed by
exogenous factors attribute a much more significant role for endogenous
forces. In that context, a suite of models and arguments that broadly
convey the new growth theory have been directed towards endogenous
factors and processes (see, for example, Johansson et al., 2001).
Those developments are of great interest to regional development ana-
lysts and also to practitioners for several reasons, including the recogni-
tion of the importance of cities and regions in the development process,
and also because they introduce an explicit spatial variable into economic
development and growth theory, which was a mostly ignored element
in neoclassical thinking. This evolutionary development is particularly
significant as the importance of regions in national economies and in
particular the role of many of the worlds mega-city regions has changed
considerably since the 1970s as a result of globalization, deregulation and
structural change and adjustment. Understanding these recently recog-
nized processes of change are crucial for analysing and understanding dif-
ferent patterns of regional economic performance and in formulating and
implementing regional economic development planning strategy.
This book focuses explicitly on endogenous regional development,
attempting to provide a range of disciplinary perspectives on the nature
of endogeneity and what endogenous regional development is about, how
it might be conceptualized, measured and modelled, and how empirical
analysis taking an endogenous perspective may be conducted at different
levels of spatial scale.
Changing Paradigms
many regions are not re-equipping themselves fast enough to compete effectively
in the global age of business and technology of the post-industrial economy.
To compete successfully in the global economy, regional organizations and
During the 1980s by which time the focus in economic policy paradigms
had shifted from a focus on Keynesian thought and the focus in regional
development thinking on the Solow (1956) model emphasizing exogenous
forces, to a new paradigm represented by monetarism and economic
rationalist thought there had been a shift from concerns about develop-
ing a regional comparative advantage to developing a regional competitive
advantage. There had also been a shift in regional development planning
strategy from master planning and structural planning to strategic plan-
ning paradigms. Thus a new way of conceptualizing regional economic
growth and development had begun to emerge, which today is known as
the new growth theory.
As early as the late 1970s, Rees (1979) had proposed that technology
was a prime driver in regional economic development, and since then
over the ensuing two to three decades the regional science literature
has shown how technology is directly related to traditional concepts of
agglomeration economies in regional economic development. Economists
such as Romer (1986, 1990), Lucas (1988), Barro (1990), Rebelo (1991),
Grossman and Helpman (1991) and Arthur (1994) sought to explain tech-
nical progress in its role as a generator of economic development as an
endogenous effect rather than accepting the neoclassical view of long-term
growth being due only to exogenous factors. In macroeconomic models
of endogenous growth, technological progress was mainly seen as an
endogenous process in an economic system, where knowledge is generally
embedded in human capital that is enhanced through education, training,
creativity and R&D. Thomas (1975) and later Erickson (1994), among
others, showed how technological change was related to the competitive-
ness of regions. Norton and Rees (1979), Erickson and Leinbach (1979)
and Markusen (1985) showed how the product cycle, when incorporated
into a spatial setting, demonstrated that some regions could be seen as
the innovators, while others become the branch plants or recipients of the
innovation, and those might even then become innovators via endogenous
growth. The concept of innovative milieu was formulated to explain the
how, when and why of new technology generation. That notion linked
back to the importance of agglomeration economies and localization
economies that had been viewed as leading to the development of new
industrial spaces (Scott, 1988; Porter, 1990). In particular, research by
Krugman (1991, 1995, 1996) and Krugman and Venables (1996) led to a
greater emphasis on knowledge as a tacit and primarily local good and the
recognition of it as a driving endogenous self-reinforcing mechanism for
regional development.
There has also been an emphasis on the importance of investment in
human capital and its role in regional development, as demonstrated in
work by the OECD (2000, 2001). In addition, writers such as Fukuyama
(1995) suggest that it is not just economic but also value and cultural
factors including social capital and trust that are important in the rise
of technology agglomerations as seen in the Silicon Valley phenomenon,
where collaboration among small and medium-size enterprises through
networks and alliances and links with universities forge a powerful R&D
and entrepreneurial business climate. How to enhance regional eco-
nomic development through engendering regional collaborative advan-
tage (Huxam, 1996) began to be seen as a desirable component of regional
development planning strategy.
There has emerged as well a considerable emphasis on the role of leader-
ship and institutions as factors that can enhance or even act as a catalytic
effect in endogenous regional development, as demonstrated by Stimson
and Stough (2009). As Rees (2001) has pointed out, technology-based
theories of regional economic development need to incorporate the role of
entrepreneurship and leadership, particularly as factors in the endogenous
growth of regions, and it is the link between the role of technology change
and leadership that can lead to the growth of new industrial regions and to
the regeneration of older ones (p. 107).
Especially in the period since the mid-1970s, the processes of globaliza-
tion have resulted in the emergence of an increasingly borderless economic
world with increasingly unrestricted mobility of capital and labour and
increasing freedom of trade in merchandise and services. Seemingly, the
influence of the nation state was reduced in a world where cities and par-
ticularly mega-city regions assumed increasing importance as strategic
hubs and as the drivers of creativity, innovation and entrepreneurial activ-
ity and as they increasingly became the dominant engines of economic
growth (Knight and Gappert, 1989; Ohmae, 1995; Prudhomme, 1995;
Florida, 2002). That created new stresses for both nations and for regions
and their governments in developing strategies to find a competitive edge
PC
EC HC
SID
CC SC
First there are five chapters in which scholars from different disciplinary
backgrounds provide their take on endogenous regional development:
from an economist, a geographer, a sociologist, a planner and a behav-
ioural/management scientist.
Kenneth Button provides an economists viewpoint (Chapter 2) in
which he refers to the powerful tool-kit of powerful techniques economists
have for analysing and understanding why some countries and regions
grow faster than others. He refers to a world without endogenous growth
and the well-known exogenous growth models in neoclassical economics
as illustrated by the Solow model before turning to discuss the emer-
gence of the new or endogenous growth theory from the 1980s, while
also reminding us of important precursors to the new growth theories
The second theme that is addressed in this book relates to the measure-
ment and modelling of endogenous regional growth and development.
Somewhat surprisingly there are few examples of models that have been
explicitly developed and empirically tested to actually measure endogen-
ous regional growth or to investigate the determinants of spatial variations
in regional performance on an endogenous growth measure.
A Way Ahead?
In the final chapter (Chapter 15) Peter Nijkamp and Roberta Capello
point out that regional growth theory covers a little more than half a
century and during that time we have deepened our knowledge and
understanding of the complex backgrounds and impediments to balanced
regional development. In that context endogenous considerations are of
great importance.
Nijkamp and Capello provide a provocative review of approaches to
investigate regional development, which begins with a discussion of the
vexed question of convergence hypothesis in the debate on regional dis-
parities. In particular they draw attention to the policy issues involved,
referring to the need for regions to achieve a role in the international
division of labour and maintain it over time, and pointing to the implica-
tions of regional growth trajectories in coping with sustainability issues
and scarcity of environmental resources, such as environmental quality,
natural resource security, urban quality of life, or spatial biodiversity. The
authors debate the need for theories of regional growth and development
to be more realistic and to better incorporate the complex and interac-
tive behaviour and processes that take place in space and to understand
regional competitiveness in terms of endogenous factors. In particular
they refer to the importance of the spatial distribution of knowledge and
REFERENCES
Acs, Z.J., Anselin, L. and Varga, A. (2002): Patents and Innovation Counts as
Measures of Regional Production of New Knowledge, Research Policy, 31(7),
106985.
Arthur, W.B. (1994): Increasing Returns and Path Dependency in the Economy,
University of Michigan Press, Ann Arbor.
Barro, R.J. (1990): Government Spending in a Simple Model of Endogenous
Growth, Journal of Political Economy, 98(5), S103S125.
Blakely, E.J. (1994): Planning Local Economic Development: Theory and Practice,
2nd edition, Sage Publications, Thousand Oaks, CA.
Capello, R. (2009): Spatial Spillovers and Regional Growth: A Cognitive
Approach, European Planning Studies, 17(5), 64360.
Capello, R. and Nijkamp, P. (eds) (2009): Handbook of Regional Growth and
Development Theories, Edward Elgar, Cheltenham, UK and Northampton,
MA, USA.
Capello, R., Nijkamp, P. and Pepping, G. (1999): Sustainable Cities and Energy
Policies, Springer-Verlag, Berlin.
Dring, Th. and Schnellenback, J. (2006): What Do We Know About Geographical
Knowledge Spillover and Economic Growth?, Regional Studies, 40(3), 37595.
Erickson, R.A. (1994): Technology, Industrial Restructuring and Regional
Development, Growth and Change, 25(3), 35379.
Erickson, R.A. and Leinbach, T. (1979): Characteristics of Branch Plants
Attracted to Non Metropolitan Areas, in: Lonsdale, R. and Seyter, H.L. (eds):
Non Metropolitan Industrialization, Winston, Washington, DC.
Etzkowitz, H. and Leydesdorff, L. (1996): The Future Location of Research:
A Triple Helix of UniversityIndustryGovernment Relations, II, EAAST
Review, 15(4), 2025.
Etzkowitz, H. and Leydesdorff, L. (1997): Universities and the Global Knowledge
Economy, a Triple Helix of UniversityIndustryGovernment, Pinter, London.
Etzkowitz, H. and Leydesdorff, L. (2000): The Dynamics of Innovation: From
National Systems and Mode 2 to a Triple Helix of UniversityIndustry
Government Relations, Research Policy, 29(2), 10929.
Florida, R. (2002): The Rise of the Creative Class: And How Its Transforming
Work, Leisure, Community and Everyday Life, Basic Books, New York.
Fukuyama, F. (1995), Trust: The Social Virtues and Creation of Prosperity, Free
Press, New York.
Garlick, S., Taylor, M. and Plummer, P. (2006): An Enterprising Approach to
Regional Growth: The Role of VET in Regional Development, NCVER (National
Centre for Vocational Education Research), Australian National Training
Authority, Adelaide.
Grossman, G.M. and Helpman, E. (1991): Innovation and Growth in the Global
Economy, MIT Press, Cambridge, MA.
Hertz, N. (2009): Cooperation Must Rule, Nature, 457(7232), 9623.
Huxham, C. (1996): Collaborative Advantage, Sage, Thousand Oaks, CA.
Johansson, B., Karlsson, R. and Stough, R.R. (eds) (2001): Theories of Endogenous
Regional Growth, Springer, Berlin.
Keeble, D. and Wilkinson, F. (1999): Collective Learning and Knowledge
Development in the Evolution of Regional Clusters of High-technology SMEs
in Europe, Regional Studies, 33(4), 295303.
Knight, R.V. and Gappert, G. (eds) (1989): Cities in a Global Society, Sage
Publications, Newbury Park.
Krugman, P. (1991): Geography and Trade, MIT Press, Cambridge, MA.
Krugman, P. (1995): Development, Geography and Economic Theory, MIT Press,
Cambridge, MA.
Krugman, P. (1996): The Self-Organising Economy, Blackwell, Oxford.
Krugman, P. and Venables, A. (1996): Integration, Specialization and
Adjustment, European Economic Review, 40(35), 95967.
Lucas, R.E. (1988), On the Mechanics of Economic Development, Journal of
Monetary Economics, 22(1), 342.
Markusen, A. (1985): Profit Cycles, Oligopoly and Regional Development, MIT
Press, Cambridge, MA.
Nijkamp, P. (forthcoming): E Pluribus Unum, Region Direct.
Nijkamp, P. and van Hemert, P. (forthcoming): Knowledge Infrastructure and
Regional Growth, Regional Symbiosis.
Nijkamp, P., Vleugel, J., Maggi, R. and Masser, I. (1994): Missing Transport
Networks in Europe, Ashgate, Aldershot.
Norton, R.D. and Rees, J. (1979): The Product Cycle and the Decentralization of
North American Manufacturing, Regional Studies, 13(2), 14151.
OECD (2000): Links between Policy and Growth: Cross-country Evidence, draft
paper for Working Party 1, Economics Department, OECD.
OECD (2001): The Well-Being of Nations: The Role of Human and Social Capital,
Centre for Educational Research and Innovation (CERI), OECD Publishing,
Paris.
Ohmae, K. (1995): The End of the Nation State: The Rise of Regional Economics,
Free Press, New York.
Porter, M.E. (1990): The Competitive Advantage of Nations, MacMillan, New York.
Prudhomme, R. (1995): On the Economic Role of Cities, in Cities and the New
Global Economy, papers from an international conference presented by the
OECD and the Australian Government, Australian Government Publishing
Service, Canberra, 3, 72842.
Rebelo, S. (1991): Long Run Policy Analysis and Long Run Growth, Journal of
Political Economy, 98(5), S71S102.
Rees, J. (1979): State Technology Programs and Industry Experience in the USA,
Review of Urban and Regional Development Studies, 3(1), 3959.
Rees, J. (2001): Technology and Regional Development: Theory Revisited, in:
Johansson, B., Karlsson, Ch., Stough, R.R. (eds): Theories of Endogenous
Regional Growth, Springer-Verlag, Heidelberg, pp. 94110.
Romer, P.M. (1986): Increasing Returns and Long-run Growth, Journal of
Political Economy, 94(5), 100237.
Romer, P.M. (1990): Endogenous Technological Change, Journal of Political
Economy, 98(5), S71S102.
Scott, A.J. (1988): New Industrial Spaces: Flexible Production Organization and
Regional Development in North America and Western Europe, Pion, London.
Simmie, J. (ed.) (1997): Innovation, Networks and Learning Regions?, Jessica
Kingsley, London.
Solow, R.M. (1956): A Contribution to the Theory of Economic Growth,
Quarterly Journal of Economics, 70(1), 6594.
Solow, R.M. (2000): Growth Theory: An Exposition, Oxford University Press, New
York.
Stimson, R.J. and Stough, R.R. (with Salazar, M.) (2009): Leadership and
Institutions in Regional Endogenous Development, Edward Elgar, Cheltenham,
UK and Northampton, MA, USA.
Stimson, R.J., Stough, R.R. and Roberts, B.H. (2006): Regional Economic
Development: Analysis and Planning Strategy, revised edition, Springer, Berlin.
Taylor, I. (2009): Learn to Convince Politicians, Nature, 457(7232), 9589.
Thomas, M.D. (1975): Growth Pole Theory, Technological Change and Regional
Economic Growth, Papers of the Regional Science Association, 34(1), 325.
There must be a purely economic theory of economic change which does not
merely rely on external factors propelling the economic system from one equi-
librium to another. (J. Schumpeter, Preface to the Japanese edition of Theorie
der Wirtschaftlichen Entwicklung, 1911, transl. 1934)
Economic progress, in capitalist society, means turmoil. And, ... in this turmoil
competition works in a manner completely different from the way it would
work in a stationary process, however perfectly competitive. Possibilities for
gains to be reaped by producing old things more cheaply are constantly materi-
alizing and calling for new investments. These new products and new methods
compete with the old methods not on equal terms but at a decisive advantage
that may mean death to the latter. (J. Schumpeter, Capitalism, Socialism, and
Democracy, 1942)
INTRODUCTION
20
has always been at the core of modern economics. Adam Smith, David
Ricardo, Malthus and other early classical economists offered a variety
of explanations for improving economic efficiency, focusing on the ben-
efits of the division of labor in one way or another, but they were largely
focused on squeezing more production from a given resource base than
economic growth per se.2 Their concern was essentially with what we now
consider comparative statics rather than dynamics.
More recently the importance of technological progress in economic
growth has attracted attention. Initially, and following the pioneering
work in the 1920s of Frank Ramsey (1928), strenuous intellectual efforts
have been made to incorporate into economic models the changes in tech-
nology that have manifestly affected economic growth since the Industrial
Revolution. The basic point being that capital and labor augmenting
technology continually increases returns to factors of production, leading
to additional investment, allowing per capita output to increase seemingly
indefinitely. The challenge has been to understand where this technical
progress comes from, how it becomes embedded in production processes
and how it will, in quantitative terms, influence the economic growth rates
of different regions. While this has largely been treated as a question of
macroeconomic interest, there are important spatial implications espe-
cially because technological progress is not ubiquitous or cannot always
be easily and quickly transferred between regions.
In the 1960s the UKs Royal Economic Society and the American Economics
Association commissioned a series of survey papers by leading figures in their
fields. Two of these, that by John Meyer (1963) on Regional Economics,
and by Frank Hahn and Robin Matthews (1964) on The Theory of
Economic Growth provide a useful basis for considering how and why
endogenous growth theory has developed over the past 45 years or so.
While there was concern at the time amongst regional economists over
spatially differential growth patterns, the emphasis of academic work, and
possibly reflecting the dominance of Keynesian economics at the time,
was more on short-term considerations. In particular there was interest
in the tracing through via multiplier and input-output analysis the local
implications of expanding a regions export base.
The neoclassical world of macroeconomic growth theory before the
1980s thought that growth was the result of forces that impinged from
the outside rather than being a product of the economic system itself.
Per capita income growth was seen as dependent on saving rates and on
exogenous shocks to the steady-state growth path that resulted.
The empirical work of the time focused on extracting the inputs of the
factors of production de facto, labor inputs from the growth process
and then treated differences in national or regional growth as an account-
ing residual. To be fair, however, in part this was as much due to data
weaknesses as to intellectual understanding.3 As we see below, endogenous
growth theory goes beyond this and seeks to look at the private and public
sector choices that cause these residuals to vary.
Comparative Statics
Region A Region A
Region B Region B
Classical Model New GrowthTheory Model
path, either up or down. However, Solow, Swan and others argued that
the capitaloutput ratio of the Harrod-Domar model should not be
regarded as exogenous. They proposed a growth model where the capital
output ratio was precisely the adjusting variable that would lead a system
back to its steady-state growth path.
The key assumption of the neoclassical economic growth model is that
capital is subject to diminishing returns. Given a fixed stock of labor, the
impact on economic output of the last unit of capital accumulated will
always be less than the one before. Assuming for simplicity no techno-
logical progress or labor force growth, diminishing returns implies that
at some point the amount of new capital produced is only just enough to
make up for the amount of existing capital lost due to depreciation. At this
point, because of the assumptions of no technological progress or labor
force growth, the economy ceases to grow.
Assuming non-zero rates of labor growth complicates matters at the
regional level, where migration may be large, somewhat, but the basic
logic still applies in the short run the rate of growth slows as diminishing
returns take effect and the economy converges to a constant steady-state
rate of growth (that is, no economic growth per capita).7
Including non-zero technological progress is very similar to the assump-
tion of non-zero workforce growth: a new steady state is reached with con-
stant output per worker-hour required for a unit of output. However, in
this case, per capita output is growing at the rate of technological progress
in the steady state (that is, the rate of productivity growth).
More formally, an exogenous growth model can be formulated by
taking a simple Cobb-Douglas production function of the form, Y = Y =
A(t)K1b Lb (where Y is net national product, K is the capital stock, L is the
labor stock and A is the level of technology). The fact that A is a function
of time (t) indicates the standard neoclassical assumption that technology
only improves over time for reasons external to the model.
Y 5 AKaL1 2a
where there are constant returns to scale for all inputs together (i.e., a 1
(1 2 a) 5 1). Therefore, it seems as if output per capita, and with it con-
sumption per capita, does not grow unless the exogenous factor, A, grows
too.
To indigenize A, the Cobb-Douglas production function for each
individual firm is defined as:
Yi 5 AiKai L1i 2a
Yi 5 G 2zK ai L1i 2a
Y 5 K a 1 zL1 2a
a social planner because firms are not, individually, taking the externality
into account. In other words, with full information and an effective tool-
kit there is a role for public policy.
Learning-by-doing within a firm means current unit costs are a function
of experience (as measured by the firms cumulative past output). Given
the learning curve for a single firm, then imitation of successful firms
on the part of other firms in the industry spreads the learning around,
such that the industry as a whole can benefit from falling-forward supply
curves. The process links unit costs to cumulative industrial output within
a country or region. The ease of imitation and learning then increases
within spatial agglomerations, which in turn can be understood as what
are often called little nations benefiting from increasing returns.
Romer refines his work by looking at why private firms directly invest in
technical knowledge by undertaking R&D when it has the characteristics
of a quasi-public good; knowledge only being partially excludable means
that the firm will not reap the full benefits. Patents, trademarks and copy-
rights are assumed by Romer, as they had been in previous work by Joseph
Schumpeter (1942),14 to protect, at least for a time, part of the knowledge
acquired, but even without them a firm generally enjoys a first mover
advantage and can costlessly reproduce for sale or internal use a new
design. It can sell these new products for a time at prices above marginal
cost and earn super-normal profits, but even as the short-term monopoly
position is eroded there is the incentive to further innovate because of the
lowered R&D costs that stem from the previous experience.15 In effect,
R&D activities become a barrier to entry of competition into the market.
Robert Lucas (1988, 1993) takes a somewhat different approach to the
spillovers from investment, focusing on education and on-the-job training.
Essentially, the interest is more on human capital rather than in physical
capital and R&D as in Romers framework.
One approach followed by Lucas involves assuming that personal
human capital is built up by workers through their postponing current
consumption in the hope that education will allow for greater consump-
tion in the future. To develop an endogenous growth effect from this, he
assumes that there is an externality from this activity because the actions
of these individuals will enhance the overall productivity of the labor
force, including those that do not defer consumption. Lucas found empiri-
cal support for this in the spatial agglomeration that can be observed,
especially in cities.
A second approach is to assume that knowledge is obtained on the job
rather than through explicit formal educational programs. Hence, labor
does not leave the workforce to acquire and develop knowledge, but gains
it from having to produce an ever-increasing range of more sophisticated
his initial model, and the lack of appropriate on-the-job training in the
second.19
Equally, capital does not move from region A to B because of the higher
returns and less uncertainty that are to be found in regions that already
have a high level of prosperity and a pool of complementary labor. While
the earlier regional models of Kaldor, in particular, focused largely on
divergent growth rates in the context of traditional-style manufacturing
industries that still dominated Western economies, the more contem-
porary form of the theory pays particular attention to the endogenous
growth that occurs in regions that have an established high-skilled work-
force, and the ability to further develop their knowledge-based industries.
Again, there is no substantive difference in the modeling whether it is con-
ducted at the national level or for smaller regions, although institutional
and cultural factors may influence the speed and detail of the growth paths
of either.
POLICY IMPLICATIONS
neoclassical model does allow for fairly limited public policy interventions
that would de facto lubricate the system and facilitate a more rapid move
to a steady-state growth path. It would not, however, produce movement
along it or shift it. In the context of migration, for example, this may
involve improved information and enhanced transportation services to
allow existing resources to migrate and be used more effectively along
Adam Smiths lines of argument of greater divisions of labor.
If there is endogeneity in the growth process then the policy options are
somewhat wider. Since knowledge is important, then diffusion of ideas
and broader national policies for R&D can be deployed to bring lagging
regions up to the production frontier enjoyed by the leading regions. To
stimulate a nations growth, Romer, for example, argues for a reduction in
the USs federal deficit to reduce interest rates that would in turn increase
the amount of human capital devoted to R&D by raising the discounted
value of any given stream of future revenues associated with a new design.
The Romer framework would also suggest subsidies for R&D because of
the currently uncompensated external benefits that it generates; in contrast
the Lucas models suggest that these subsidies should largely go to the
education and training of workers.25
There is also a case for freer trade in that it allows for the more rapid
diffusion of knowledge and thus breaks down the monopoly of those
regions and countries that currently enjoy its ownership. More generally,
it releases knowledge workers to invent new designs rather than for those
in the lagging regions having to expend energies on catching up and effec-
tively continually having to reinvent the wheel. If one pursues the Richard
Florida (2005) line of argument that the creative classes are attracted
and retained by the larger environment in which they live and work, then
investment in various forms of social and economic infrastructure become
important.26
CONCLUSIONS
is still far from complete. In part this is because the economic modeling
remains controversial and this in turn can partly be explained by gaps in
the work that has been completed by complementary disciplines.
While there are now relatively powerful economic and programming
techniques available to examine macro- and meso-data our understanding
of some of the fundamental micro-forces that lead to economic growth is
still relatively poor.27 To more completely understand the motivations for
companies to undertake R&D, and for individuals to invest differentially
in enhancing their knowledge bases, requires more systematic analysis by,
in the first case, management scientists, and in the latter by sociologists
and social psychologists. Even at the more macro level there appear to be
fundamental cultural differences that affect the speed and nature of uptake
of new technologies, and especially those that emerge in other regions that
are not normally seen as within the scope of economics. Similarly, there
appear to be variations in the flexibility to which nations or regions can
deal with shocks to their economic systems even within what superficially
appear to be relatively similarly educated populations (Pack, 1994). As the
old saying goes, The devil is in the detail, and useful analyses of many
of these details require a broader set of tools than those conventionally
deployed by traditional economists.
NOTES
Y = g(L, K, T)
change arising from profit-squeezes or, in the particular case of Smith, arising because
of previous technical conditions. This was understandable. At the time of his writing,
the UK economy was moving into an extended period when economic growth reached
about 1 percent per annum compared with the 0.5 percent per decade that had previ-
ously been the norm (i.e., effectively peoples living standards did not change over their
life-times) and Smith was, in a way, groping to understand this new world.
3. Harry Richardson (1978, p. 18) says of the empirical work on regional growth up to the
1980s that, The models developed so far have been handicapped by severe data limita-
tions, too few observations, the reliance on static models, their recursive structure, the
heavy dependence on national economic variables, and the neglect of space.
4. Because of its market-clearing assumptions the pure classical model would be driven
only be income differentials.
5. Strictly, with full, instantaneous market clearing there is no unemployment in this type
of model, labor movements being determined by real relative wages. The unemploy-
ment effect is added to indicate possible imperfections in the short-term labor markets
in the two regions.
6. This model synthesizes the work of Evsey Domar (1947) and Roy Harrod (1939).
7. The Malthusian model, because of its implied capital productivity function, may be
seen as an extreme case of this whereby society cannot rise above the subsistence level
of per capita income.
8. McCraw (2007), for example, argues that part of Schumpeters motivation for the style
of presentation in the book was due to a degree of chagrin felt over the popular success
of John Maynard Keynes General Theory that had overshadowed his own analysis of
trade cycles. Strict modeling along Schumpeterian lines did not emerge until the 1990s.
9. Another of the stimuli for the interest in developing a more complete understanding
of economic growth in the 1950s and 1960s was institutional and founded on concerns
about the economic performance of the economies of the newly independent, ex-
colonial nations and the role that institutions such as the World Bank should play in
assisting them. This was at a time when it became clear that developing countries were
not catching up with higher-income nations.
10. For Solows later thoughts on the roles between endogenous and exogenous growth
theory see Solow (1994).
11. Romers assumption that knowledge and physical capital are related stemmed from
empirical observations that there were greater correlations between investment and
output than the neoclassical model would predict.
12. Romer (1990b) provides a series of anecdotal examples to support this view.
13. How the stock of knowledge is to be measured is a moot point (Steedman, 2001). There
are various elements or components to knowledge that need weighting to determine
the stock, even if they can be defined and measured. There are also questions as to the
extent that knowledge is non-rival rather than being unique to individuals.
14. Joseph Schumpeter provided a general model of economic growth in which through
the forces of creative destruction in oligopolist markets, entrepreneurs were stimulated
to develop new technologies. His discussion of endogenous growth was couched very
much in terms of a critique of Karl Marx but now emerged into a rigorous model of the
type Romer and others later developed.
15. This raises issue of whether endogenous growth is, in effect, an assumption in the model
or a result of the model.
16. The model is not closed, however, because Lucas only assumes that new goods are
continually being produced without having an explicit model of why.
17. Studies of economic convergence provide a pragmatic way of indirectly testing endog-
enous growth theory without the need to specifically define and measure knowledge.
18. For a much more rigorous analysis of endogenous growth involving two regions or
countries see, Segerstrom et al. (1990).
19. Richard Florida (2005) has added to these ideas by arguing that creative classes having
preference for where they live, and that the attributes required to attract and retain
them are more likely found in regions with an established high-income and educated
labor force. In either the older or more recent formulations of this framework, the
implications are often circular-and-cumulative causation; essentially richer regions get
richer and poorer regions, poorer.
20. Theoretical divergence is not an automatic outcome of endogenous growth (Kelly,
1992). Robert Tamura (1991), for example, by making the simple assumption that an
idea is more difficult to discover than it is to learn, and following from this that there is
a higher payoff from education for people with a lower basic level of education, dem-
onstrates that convergence can occur. Put another way, if a person is at the technology
frontier, movement out requires new knowledge whereas for someone within the fron-
tier it is possible, and easier, to move out by acquiring existing knowledge.
21. This approach essentially looks for changes in standard statistical indicators of
dispersion normally the variance.
22. A standard estimating equation takes the form D(yit) = a + b(yitj) + DiG + e it, where
the yit is that being tested for convergence, Di is a (n k) matrix of k variables capturing
spatial economic heterogeneity, G is a (k 1) vector of parameters, a and b are constants
and eit is a white noise error term. The condition b > 0 is a necessary but not sufficient
condition for b-convergence (Bernard and Durlauf, 1996). A critique of the empirical
work using this and similar tests for convergence is provided by Howard Pack (1994).
23. Bradford De Long and Lawrence Summers (1991) make the general point in their
specific analysis of investment in equipment, to which they credit much of the positive
spillover effects on economic growth, that markets are not perfect and that bad policy
decisions (government failures) inevitably occur and will influence the nature of the
growth path for any economy. Some studies seek to embrace such effects in the analysis,
for example, Barro (1991) includes political stability, potential rent seeking, and other
political variables in his calculations while Button and Pentecost (1995) embrace mon-
etary policy within their work on the European Union.
24. Button and Pentecosts (1995, 1999) work on the regional economies of the EU find
that there is evidence of divergence when no allowance is made for possible national
steady-state growth path differences. The national boundaries used by Barro and Sala-
i-Martin would not seem to correspond to natural economic units.
25. There would seem to be another policy difference in terms of the types of industry that
central policy-makers may seek to encourage to locate in economically slow-growing
regions. The tradition of seeking any form of economic activity to stimulate Keynesian
multiplier effects is alien to both approaches. Romers framework suggests that there is
a need to locate more knowledge-based elements of firms and industry in such regions
to stop their cumulative decline. Lucas, in contrast, because of the perceived impor-
tance of learning-by-doing, may well argue that provided industry achieved this it
would set in motion an effect to stimulate faster long-term growth.
26. The evidence that investment in more traditional infrastructure in slow-growing regions
can stem the tide of divergence is less certain (Button, 1998).
27. This is not to say that these techniques are always fully exploited. As Pack (1994)
argues, there are likely strong interactive effects between the factors that lead to eco-
nomic growth but there is a tendency for econometric analysis to miss these interactions
in the specifications used.
REFERENCES
Abramovitz, M. (1986), Catching up, forging ahead, and falling behind, Journal of
Economic History, 46(2), 385406.
Arrow, K.J. (1962), The economic implications of learning by doing, Review of
Economic Studies, 29(3), 15573.
INTRODUCTION
The question posed in this chapter is: How do we understand and inter-
vene in processes of regional economic growth and change?
Globalization is eroding the old economic certainties, increasing com-
petition, reconfiguring the demand for skills, forcing the pace of tech-
nological and managerial change, transforming inter-firm relationships
and changing (through commodification) the role of finance. Some
would argue that there has been a sea change in the workings of national,
regional and local economies with the end of Fordism and the coming of
post-Fordist flexible specialization and mass customization. For policy-
makers, the pressing issue is how can appropriate and workable policies
and programmes be developed to cope with the internationalization of
production and consumption leading to rapid and radical change, foster
sustainable and robust local growth and enhance local capacities beyond
providing short-term palliatives for unemployment and its associated
social problems?
From an economic geography perspective, two bodies of discourse
currently dominate our understanding of the processes driving local
economic change: (1) the discourse of endogenous growth in the new
economic geography of the economists; (2) the new regionalism dis-
course along with its social constructionist and relational underpinnings in
economic geography proper and equivalent areas in other social sciences.
Based upon seemingly contradictory criteria of success, derived from
either the stylized facts of endogenous growth theory or the contingency
of new regionalism, these bodies of discourse have become reified into a
discourse of policy and practice based upon stereotypical understandings
of the mechanisms that are supposed to characterize the development of
local economies in a world of globalization. Here we argue that, despite
39
et al., 1998; Porter, 1998; MacKinnon et al., 2002) now more generally
captured under the term new regionalism (Rainnie and Grobbelaar,
2004). Building directly on the ideas of Granovetter, the new regionalist
writing suggests that economic growth is linked not only to market condi-
tions, but also to repeated inter-firm interaction and knowledge exchange,
collaborative long-term buyersupplier relationships, the creation of
social capital (including trust, reciprocity and loyalty) and a supportive
tissue of local institutional thickness (see Putnam, 1993; Cumbers et al.,
2003; Malmberg and Maskell, 2006).
The new regionalism and embeddedness ideas have prompted renewed
interest in the advantages of geographical proximity between firms in
related industries (Keeble and Nachum, 2002) and it is a significant shift
away from traditional understandings of agglomeration based on input
output relations and transaction costs, focusing instead on the social and
institutional drivers of growth. Whilst the approach has theorized new
drivers and has added richness to the local growth debate, it has been less
rigorous in how we might measure those processes or detect their impact
on long-run economic growth. It is also a model of economic growth with
great policy appeal, but the application of this new orthodoxy has often
been implemented uncritically and in the absence of detailed empirical
evaluation (i.e., evidence-based policy).
The new regionalism approach may be popular but it also has sig-
nificant and prominent limitations (see also Taylor, 2005). First and
most importantly, the mechanisms of local economic growth transmis-
sion are not clearly articulated in the model, and this limitation has four
interconnected components:
example. However, as more theorizing has been added to the model, it has
become evident that there is no mechanism to say what the importance of
the separate, postulated processes is in promoting or retarding regional
economic growth. It is here that the modelling strategy of endogenous
regional theory has much to offer.
The Garlick et al. (2007) study of vocational education and training (VET)
provision in Australia focused on providing advice on policy alternatives
that might better target VET to foster regional economic growth that was
consistent with the long-run processes shaping those regional economies.
The study asked two key questions: (1) What processes are driving eco-
nomic change at the regional scale in Australia? (2) How can VET work
sympathetically and in a targeted manner to modify those processes, create
growth and, therefore, more fully realize national economic potentials?
From an institutionalist perspective, successful local economies are said
to depend on complex processes of integration and embedding built on
trust, reciprocity and loyalty that create social capital. They involve
the exchange of information, ideas and innovation through mechanisms
of quasi-integration that creates new knowledge through processes of
situated learning. In addition, the support of local institutions (labelled as
institutional thickness; Amin and Thrift, 1994) is seen as bolstering these
local economic processes. Enmeshed in webs of global coordination and
value transfer, these local economies are seen as nodes of untraded inter-
dependencies (Storper, 1997) that are tapped into by large corporations
and TNCs, which, for their part, act as global information arbitrageurs
(Economist, 2003).
To reflect institutionalist thinking on the propulsive elements of local
economic growth, eight drivers were distilled from the broad range of
theories, each of which prioritized different sub-sets of drivers as operating
in different ways, depending on local context. These were:
The eight drivers distilled from theory were measured empirically using
a unique Australian data set generated across Australia between 1984
and 2002, with the country divided into 94 functional regions. The nature
and origins of the data set are outlined in Appendix 3A.1 at the end of
this chapter. These empirical measures were incorporated into a conven-
tional gap convergence (mean reversion) econometric model (Plummer
and Taylor, 2001b; Taylor and Plummer, 2003) in which regional growth
(measured in terms of unemployment relativities) is broken down into
three components:
Very clearly, human capital and enterprise are the life blood of regional
economic growth, especially when there is appropriate government
intervention.
However, the study recognized that not only must positive drivers of
growth be found at the regional scale as the quantitative modelling shows,
but, to achieve success, it is also important that growth transmission pro-
cesses at that scale are faced by as few impediments as possible. To address
this aspect of the local growth process, qualitative information was
tendency to argue that, of course, things had changed and that the future
promised growth. This could only be interpreted as the optimism of the
immediate. It highlights, however, the potential limitations of qualitative
analyses and, possibly, the unreal expectation that consultants can create.
Not all regions were the same, and in some outstanding initiatives had
been taken. However, where plans were developed, there was a tendency
for people to be transfixed by the plan and not to move to implementa-
tion. It was concluded that 20 years of bottom-up regional development
policy and practice in Australia had produced disturbingly low levels of
local action and that government-mandated regional leadership might
not be the most effective way to encourage local growth that depends on
free-flowing ideas and enthusiasm across the whole of a region (Garlick et
al., 2007).
On a more positive note, the workshops did recognize a range of
impediments to growth transmission. In particular, they recognized a
general complacency and lack of dynamism in some regions. They pointed
to human capital being exported to major metropolitan regions across the
country, with this exodus being seen as normal. Key industries were not
seen as producing business spin-outs at the regional level, and that facili-
tative regional planning to realize opportunities appeared to be disorgan-
ized. At the same time, there was a recognition that the education system
was unable to inspire a regions human capital to be enterprising. What is
more, the workshops saw an overemphasis on specific sectoral winners,
pushed by institutions and consultants from outside the region, as the
silver bullets to achieve local economic growth rather than building on
unique local capacities, attributes and abilities (ibid.).
and jobs at least to maintain and possibly to enhance the well-being and
standard of living of entire communities that are currently facing global
competitive pressures. It involves having a strong grasp of, amongst
many other things, markets and market potentials, sourcing, financing,
business planning, accountancy, cost control, timing, coordination and
negotiation. It involves outcome-orientated people who, at their most
successful, are able to work in teams that blend complementary skills.
The enterprising outcomes of these people are not necessarily just eco-
nomic in the sense of creating new businesses. They may also emerge as
new social, cultural and environmental outcomes that reshape the very
nature of the communities within which they operate and may, in turn,
create a climate for further enterprising in that place. All the case study
regions pointed to the lack of a local enterprising culture within them
and none had mechanisms in place to facilitate the creation of such an
enterprising capacity.
Essential to the creation of enterprising human capital are mutual and
creative connections among a regions human capital what more broadly
can be called regional engagement. It is, at its most fundamental, a frame-
work of mutual cooperation and dialogue that complements and extends
the processes of enterprising human capital. However, it is more than a
process. Regional engagement is also about achieving results and generat-
ing outcomes. It is the community equivalent of the enterprising human
capital of individuals.
Towards a Policy
However, it is possible to equip those who are keen to try to take risks
with the skills that can help them succeed, including:
The reaction to these policy proposals for VET has proved very conten-
tious. It proposes a very different educational agenda from the one that has
traditionally been followed. It makes educators uncomfortable because it
disturbs peoples comfort zones and involves doing something new!
CONCLUSION
In developing a geographic perspective on endogenous processes operat-
ing in regional and local economies, this chapter has sought to highlight
the differences in methodology in this discipline compared with econom-
ics: between the intensive qualitative approach in geographys new
regionalism and the extensive modelling strategy of economics endog-
enous growth theory. Both are distinctive caricatures of local economic
processes built on sets of differently stylized facts.
It has been suggested that the two approaches are complementary and,
through integration, have the potential to bring a fuller and more nuanced
perspective on local economic growth processes and issues that are
capable of reflecting the unique characteristics and capabilities of places
rather than generating a one-size-fits-all policy prescription. As such it is
an approach that has the potential to suggest new policy initiatives.
A multi-method analysis of Australian regional change using theoreti-
cally informed empirical modelling coupled with qualitative information
collected through facilitated workshops in case study regions, threw new
light on regional growth processes in this economy. Focused on a review
of VET in Australia, this approach was shown to have the capacity to gen-
erate new policy perspectives that fall outside the usual comfort zone of
bureaucratic policy-makers. From this it might be suggested that a multi-
method approach to the analysis of endogenous processes in regional
economies might be a useful way of developing new regional policies and
programmes.
REFERENCES
Amin, A. and Thrift, N.J. (1994), Living in the Global, in A. Amin and N.J.
Thrift (eds), Globalization, Institutions and Regional Development in Europe,
Oxford University Press, Oxford, pp. 122.
Braczyk, H.J., Cooke, P. and Heidenreich, M. (eds) (1998), Regional Innovation
Systems: The Role of Governances in a Globalized World, UCL Press, London.
Brakman, S. and Heijdra, B. (2004), The Monopolisitic Competition Revolution in
Retrospect, Cambridge University Press, Cambridge, UK.
Brakman, S., Garretson, H. and Von Marrewijk, C. (2001), An Introduction to
Geographical Economics, Cambridge University Press, Cambridge, UK.
Clark, G. (1998), Stylized Facts and Close Dialogue: Methodology in Economic
Geography, Annals of the Association of American Geographers, 88(1), 7387.
Cumbers, A., MacKinnon, D. and Chapman, K. (2003), Innovation, Collaboration
and Learning in Clusters: A Study of SMEs in the Aberdeen Oil Complex,
Environment and Planning A, 35(9), 1689706.
Durlauf, S.N. and Quah. D. (1999), The New Empirics of Economic Growth,
Bureau of Statistics IRIS database. This enables the removal from the
region of establishments with the weakest local affiliations.
59
appropriated by the state to devolve not power and control but responsi-
bility and costs (see, for example, Herbert-Cheshire, 2000).
ERD has had much policy significance in the European Union (EU)
because of the realization that conventional subsidies to agriculture were
not effective in promoting growth or development in rural areas and that a
new approach focusing on local potentialities was needed (OECD, 2006).
Considerable interest in ERD occurred because of the EU LEADER
programme (discussed later in this chapter), which was conceived as an
integrated and endogenous approach to rural development (ibid., p. 90).
There is not a singular coherent position to ERD in the sociology dis-
cipline, but a range of approaches. Furthermore, there is no agreed-upon
definition (Ray, 1999; Van der Ploeg et al., 2000). Nevertheless, perhaps
Long and Van der Ploeg (1994) best sum up the sociological perspective.
For them ERD is founded mainly, though not exclusively, on locally avail-
able resources, such as the potentialities of the local ecology, labour force,
knowledge, and local patterns for linking production to consumption, etc
(pp. 12). ERD and exogenous development should be seen as a dualism of
ideal types that blend in the development strategies of regions rather than
as a mutually exclusive dichotomy (Cristovao et al., 1994; Kneafsey et al.,
2001). They are, in essence, a heuristic device (Van der Ploeg et al., 2000).
There has been some criticism of ERD along the lines that it is a nice
ideal but not practical, that local areas will never be free of external
selection of the many possible cultural assets regions can identify to make
their region special. Many regions may also have natural resource assets
that would also attract interest. Perhaps it should be noted that there has
been some critique of this approach. Graham Day (1998), for example,
argues that it shows a misunderstanding of the concept of culture in that
it focuses on the material artefact rather than on social relationships and
shared understandings, meanings and values.
Phases
Components
5. Integrated approach: the actions and projects in the local action plan
should be linked and coordinated as a coherent whole. Integration may
concern actions within a single sector, all programme actions or specific
groups of actions, or, most importantly, links between the different
economic, social, cultural, environmental actors involved in the area.
6. Networking and cooperation between areas: by facilitating the
exchange and circulation of information on rural development poli-
cies and the dissemination and transfer of good practice and innova-
tive strategies and actions, the LEADER network aims to limit the
isolation of LAGs and to create a source of information and analysis
of the actions. Some LAGs have spontaneously organized themselves
into informal networks. Another core part of LEADER is the coop-
eration between rural areas. Cooperation between areas can be trans-
national but may equally take place between areas within the same
member state.
7. Local financing and management: delegating a large proportion of
the decision-making responsibilities for funding and management to
the LAG is a key element of the LEADER approach. However, the
degree of autonomy varies considerably depending on the member
states specific mode of organization and institutional context.
Significance
CONCLUSION
NOTES
1. As I have been requested to take a sociological perspective, I will use the words sociol-
ogy, sociologist and sociological. However, I point out that a sociological perspective
is not necessarily distinguishable from a more generic social science position and cer-
tainly blends with and draws heavily on the social and human geography literature.
Further, although I have been a member of the Australian Sociological Association
for over 25 years and am a former President of the International Rural Sociology
Association, I also identify as a human geographer and am a member of the Institute of
Australian Geographers and a fellow of the Royal Geographical Society in the UK. In
general, I take an eclectic approach and ascribe to transdisciplinarity.
2. It is curious that the OECD report The New Rural Paradigm (OECD, 2006, p. 101) has
a peculiar wording of Liaison Entre Activits du Dveloppement de lEconomie Rurale
rather than what is most likely correct, Liaison Entre Actions de Dveloppement de
lEconomie Rurale, at least according to the majority of sources consulted including the
FAQ website for LEADER+ http://ec.europa.eu/agriculture/rur/leaderplus/faq_en.htm
and the French language version of the EU portal: www.welcomeurope.com (both
accessed 19 May 2010.).
REFERENCES
INTRODUCTION
73
The earliest formal literature of settlement describes how the local earth
endowed and the resources generated a rationale for placing certain infra-
structure, from roads and bridges to aqueducts, in an orderly fashion
designed to maximize resource exploitation and protect or advance some
form of settlement scheme. This set of activities evolved into what we
now term professional planning. The earliest records indicate that the
Romans, the Egyptians, the Greeks and other early civilizations had a very
high regard for laying out settlements to advance specific purposes. The
Thus, the larger and more profound aspect of indigenous resources for
planning is how land is used and controlled, ranging from parcel sizes and
heights to access and transportation. In this case the use of land for various
purposes is based on an assessment of how the in-place resources might be
used for various purposes and linked together to form new opportunities.
Planning a railroad to carry natural mineral resources is a good example.
In this case, the natural resource that is a collective resource is exploited as
a reasonable plan for current and future use. But as the exploitation takes
place other natural resources may well be jeopardized. So, the fact that
the plan is to use something indigenous is not in itself an attribute so the
planning process of collective assessment has to include all of the options,
opportunities, barriers and threats this resource requires. This is how
planning is based on the endogenous aspects of the setting.
As we enter a new millennium the local asset is no longer what the place
is but who is in the place and how they understand the use of place assets
in presenting the human knowledge base to the local region and the world.
Computer technology transforms the notion of endogenous advantage
since global interchange can potentially now take place from any loca-
tion thus making many activities completely footloose in that regard.
Until the advent of computer technology, places were valued by their
connectivity via water, rail, road or air. Now, the place as natural setting
or collection of human skills can be an endogenous value. For example,
call centres handling airline reservations around the world can be located
in a picturesque rural area that is attractive for lifestyle reasons for the
call centre workforce or because the area is less expensive but the quality
is very high. Many small rural communities in the US are now economi-
cally robust because they are attracting a cross-section of people who can
work from these locations locally and globally. In Australia, the so-called
sea-changers who move to coastal locations are founding new business in
small coastal villages and linking to the world (Gurran et al., 2007). So the
endogenous resource to be planned in this case is how to provide locally
access to the ITC infrastructure that is necessary for small- and medium-
sized businesses (including one person operations) to have access to the
web, so that a local endogenous advantage in this regard is equivalent to
that elsewhere where such infrastructure access is available.
1. It assumes that better and not more populations is the key to eco-
nomic improvement.
2. It bases economic growth not on attracting firms but attracting and
retaining high-quality human resources with the institutions, learning,
research and innovation to retain them.
3. It places all communities in the global economy and not competing
with one another within their regions for the same market share of a
diminishing resource base. It proposes multi-economic bases versus
single industrial or sector bases for all communities, both urban and
rural.
4. It views social equity as critical to community well-being and stability.
5. It places technological and social innovation at the forefront of
economic expansion. (Ibid. pp. 478)
Places that import economic activity become victims of the import rather
than the resource to build continuing wealth. Thus, it is no accident that
poverty and natural resource extraction are so closely linked because the
extraction of the natural resources usually degrades the human resources
as well. Of course, when the natural resource is gone, the human resources
are discarded. Factory towns are the prime example. In the early years, the
town has a competitive advantage with energy and or labour cost being
low. The wage, transport or resource arbitrage as it is called the differ-
ence between the costs and profit margin is reduced over time as other
places closer to the markets or with lower labour or energy cost compete
for the same goods production. As a result, the factory town dies and its
workers are stranded. The only way for places to remain competitive in
a global economy is to build on local resources and retain the wealth or
added-value components as close to the source of exploitation as possi-
ble. Similarly, manufacturing and related mass production activities as
standardized as Vernon (1966) in his classic work has shown will move
to the lowest-cost areas, leaving the last host denuded to the capacity to
adjust and use the same skilled workforce or other resources to maintain
economic perch.
The product cycle theory goes like this according to Vernon (ibid.,
p. 52):
Economic development is defined as the creation of new products and the
diffusion of standardized products. Development originates in the more devel-
oped region and is exported to the less developed region through trade and
investment. Establishing a new industry in the less developed region creates a
progressive force that can help eliminate the barriers to interregional equality.
but crafters of the future by organizing and expressing how to use local/
regional resources optimally in ways that sustain any region or community.
REFERENCES
Blakely, E.J. and Bradshaw, T.K. (2002), Planning Local Economic Development,
Sage Publications, Thousand Oaks, California.
Gurran, N., Blakely, E.J. and Squires, C. (2007), Governance Responses to Rapid
Growth in Environmentally Sensitive Areas of Coastal Australia, Coastal
Management, 35(4), 44565.
Henton, D., Melville, J. and Walesh, K. (2006), Grassroots Leaders for a New
Economy: How Civic Entrepreneurs Are Building Prosperous Communities,
Jossey Bass Nonprofit and Public Management Series, San Francisco.
Hoch, C., Dalton, L. and So, F. (eds) (2000), The Practice of Local Government
Planning (3rd edn), Washington, DC: International City/County Management
Association, Municipal Management Series.
Putnam, R.D., Feldstein, L. and Cohen, D. (2003), Better Together: Restoring the
American Community, Simon & Schuster, New York.
Stimson, R.J., Stough, R.R. and Roberts, B.H. (2002), Regional Economic
Development: Analysis and Planning, Springer, New York.
Tomlinson, R., Beauregard, R.A., Bremner, L. and Mangcu, X. (2003), Emerging
Johannesburg, Routledge, London.
Vernon, R. (1966), The Myth of Urban Problems, Harvard University Press,
Boston.
Yunus, M. (2007), Creating a World Without Poverty, Public Affairs Books,
Washington DC.
INTRODUCTION
83
underline the contribution of the present chapter. The chapter does not
claim to provide a comprehensive account of all of the factors involved in
the vibrancy of Silicon Valley; however, it does claim to elucidate a very
important factor that is systematically overlooked by the intersecting
disciplines that constitute regional economics. The chapter further claims
that this factor has been overlooked for reasons related to systematic
biases in the values, framework assumptions and organizing metaphors
through which these disciplines structure, relate to and attempt to explain
reality.
The regions technology upstarts proved more adaptive than their older, verti-
cally integrated counterparts in an environment of fast-changing markets and
technological advances. In the 1960s and 1970s, Silicon Valleys semiconduc-
tor companies outperformed older East Coast corporations like RCA and
Sylvania, and in the 1980s the regions personal computer industry outper-
formed large, established companies including NCR, Honeywell, and Digital
Equipment Corporation.
varied along three main dimensions ... each characterized by three or four
fairly distinct options or approaches (ibid.). The three main dimensions
are:
Some in the hippie milieu preferred simply to ignore the world of work, perhaps
living by their wits or the generosity of friends or parents. Some sought to rob
the system, as described how-to-do-it style in Abbie Hoffmans Steal this
Book. For many the strategy was to grudgingly coexist with the system. Get a
job, even a haircut if you must; earn the money you need to do what you have
to do, but no more. (Ibid., p. 203)
Also, these authors report that in Silicon Valley there is an ethos that gains
are shared not based on hierarchical position, as in Route 128 firms, but
instead based on contribution.
Fleming and Marx (2006) present the concept of small worlds, social
networks that cut across individual firm boundaries, in explaining how
the effective management of innovation in an environment of ongoing
knowledge exchanges was a critical factor underlying the success of Silicon
Valley. Small worlds, families of interconnected firms, are more likely to
be the source of innovation than individual firms (see also OBrien, 2007).
Saxenian (1999 and 2006) shows how this small worlds phenomenon
was also made especially effective by the many highly educated immi-
grants who came to Silicon Valley. Saxenian (1999, 2000, 2006) shows
how these families of interconnected firms went global as a result of the
social networks of immigrants/expatriates largely from Asian countries
(also reported in Adams, 2005). Saxenian documents the social networks
of Indian, Chinese, Taiwanese and other immigrant groups and how these
associations, interactions and knowledge-sharing not only substantially
advanced the development within Silicon Valley, but also created global
technological and business networks beyond what would otherwise have
been possible. In both of these contexts, domestic and international, the
immigrant communities significantly augmented the overall size of the
pie. While Route 128 firms were busy protecting their turf, Silicon Valley
firms and individuals, inspired by innovation itself and coming from a
mindset of abundance rather than scarcity, shared freely and thereby
created a better result at each level from the individual through to the
whole.
Across diverse metrics, Saxenian (2000) shows that immigrants, often
from Asian countries such as China and India, were a vital part of the
Silicon Valley technological and entrepreneurial success. For example,
approximately one-quarter of Silicon Valley technology firms founded
between 1980 and 1998 had Chinese or Indian executives (p. 252), and this
number is likely to be an underestimate because racism powered by and
blueprint have performed better over the ensuing years compared with
ventures based on the other blueprints. The Commitment blueprint high-
lights a reliance on emotional connections and family-type relationships;
Baron and Hannan (ibid.) argue that perhaps given the rarity of such
signals in the corporate world, the few companies that send out these
signals end up enjoying a competitive advantage that is hard to imitate,
particularly because it runs counter to conventional corporate ideology
in ways that are humanistic and thus that people readily find valuable.
It should perhaps not be surprising that the most successful variant, the
Commitment blueprint, is one in which people are treated and valued as
people rather than instrumentally.
In short, researchers have identified a number of unique cultural fea-
tures of the Silicon Valley region that have contributed to its ability to
bypass the more traditional high-tech corridor, Route 128. These features
include but are not limited to:
APPRECIATIVE INTELLIGENCE
Thus, in this part of the chapter we consider a quality that has been
ignored in the literature, namely appreciative intelligence, or the ability
to focus not on problems and shortcomings but on the beauty and poten-
tial inherent in any moment, including moments of suffering (e.g., to see
the mighty oak in the acorn). This is the type of intelligence that our
best school teachers have shown when they saw and brought out more
in us than we may even have known at the time that we had. In contrast
to intelligence focused on problem-solving which inevitably thereby
everywhere sees problems and solutions appreciative intelligence is a
way of being in the world that always sees the constructive side of persons
and situations without being utopian or unrealistic. It is about seeing and
taking a stand for our own and each others best selves. When we take this
stand completely and holistically, rather than partially and with limits, it
gains enormously in power and authenticity. In other words, if we see an
opportunity to make money by exploiting others, the narrowness of our
appreciative focus, and especially the lack of appreciative intelligence
towards other persons, constitutes a blind spot that will in some way feed
back around to impact our own well-being.
One of the most powerful aspects of the appreciative intelligence shown
in Silicon Valley is the very fact that it was focused on creating possibilities
that were positive intrinsically and in multiple dimensions in how people
organized work together, in the authenticity of their intrinsic enjoyment in
creating helpful and useful technologies and in their love of what they did
rather than being focused on competing with others and winning. We have
already seen several examples of this, such as the case of Google seeking
ideas from anywhere in the organization (rather than just the top) and
providing its engineers with a day a week to work on whatever personal
pet projects they wish. This practice demonstrates both how the employ-
ees were motivated by the love of discovery and problem-solving, such
that the day a week to work on their pet projects was a perk of working
at Google, and the organizational wisdom of Google that people usually
do their best work when they are doing what they most enjoy. In the end,
the organization will be enhanced from allowing people to be maximally
self-expressive, whether or not that self-expression ever yields financial or
reframing;
appreciating the positive;
seeing how the future unfolds from the present.
persistence;
conviction that ones actions matter;
tolerance for uncertainty;
irrepressible resilience.
Reframing
This is the psychological process whereby a person intentionally views or
puts into a certain perspective any object, person, context or scenario. One
of the most common examples of framing is that of calling a glass half
empty or half full (Thatchenkery and Metzker, 2006, p. 6).
We have tremendous choice in how we see situations, ourselves and
others: what we pay attention to; what we take to be important; whether
we see things as a liability or an opportunity for learning and growth, and
so on. Often life appears to us as a given reality, and older traditions
in Western social science reinforced that way of seeing (contemporary
social sciences and philosophy of science have seriously revised this view,
however). In fact, we can always choose how we make meanings out of
situations and how we respond, and people exercising their appreciative
intelligence do just that. Viktor Frankl (1963, p. 104), a concentration
camp survivor, famously wrote: everything can be taken from a man but
one thing: the last of the human freedoms to choose ones attitude in any
given set of circumstances, to choose ones own way.
Appreciating
This second element of appreciative intelligence builds on the first and is
the most constructive aspects of a situation in a realistic fashion. We can
see the power of this element most forcefully when we think about being
on the receiving end of it. If you go well out of your way to do something
nice for someone, and they focus only on flaws in the content of how you
selected to express this goal, you can see that they are missing the most
powerful and generative aspect of the situation your good intentions. If,
instead of focusing on the problems in the execution, they focused on the
beauty of your intentions and from there, in a space of alliance, requested
a different mode of expressing the intention, the experience and result of
the interaction, as well as of downstream results and interactions, would
be very different indeed.
In Silicon Valley, this positive valuing was largely present along some
dimensions and in some ways, and not in others. For example, relative to
Route 128, Silicon Valley often embraced people regardless of the superfi-
cial polish and sheen important in most traditional business environments
the pattern quoted earlier of not caring how the talent is wrapped. This
valuing sometimes also may have extended to seeking to value employ-
ees as persons rather than as means to ends, as is suggested somewhat
by the prevalent commitment blueprint of organizations described by
Baron and Hannan (2002). At the same time, this valuing was incomplete
in that immigrants faced discriminatory limitations in advancement in
many firms and from many venture capitalists. Moreover, the valuing was
somewhat limited in that, while technical insight may have functioned to
create greater equality across people who might have been discriminated
against based on other superficial dimensions, this valuing did not extend
universally to persons as persons such that people of all sorts and in all
professions (janitors, construction workers) were accorded the same levels
of social appreciation, support and opening of possibilities, along the lines
Paolo Freire sought to establish in his emancipatory educational institu-
tions (Freire, 1996). While much more common than in other parts of the
country, this valuing wasnt universal, as the example of PayPal shows,
where employment was based on being like the founders along many
superficial dimensions including aversion to sports (OBrien, 2007). At the
same time, these exceptions existed in a context in which organizations
themselves had diverse cultures and thus, across organizations, the result
would have been a valuing of a distribution of characteristics. While less
see the oak in the acorn. They also go beyond they plan[t] their acorns and
persevere to help them grow. While others may doubt the potential of the
acorns, these leaders believe in their own and others abilities to water and ferti-
lize the plants from sapling to tall oak. They deal with the risk and uncertainty
that comes with planting something new and hoping for growth. Finally, they
find a way for the oaks to survive and thrive despite unpredictable circum-
stances or a challenging environment. (Ibid., p. 33)
How does all of this relate to the Silicon Valley experience? Examples
abound along many of the dimensions (reframing, appreciating the
positive, seeing the future unfold in the present) and qualities (resilience,
persistence, conviction that ones actions matter, and comfort with ambi-
guity) of appreciative intelligence. For example, we have just encountered
elements of a general Silicon Valley culture of allowing everyone to be
who they are, regardless of how the more traditional firms might view
them as misfits, socially inept, counter-establishment, or otherwise failing
to have the outer demeanor and sheen that traditional business culture
unfortunately trains people to value. In the words of the venture capitalist
quoted earlier from Delbecq and Weiss (2000), I dont care how the talent
is wrapped. There are, as we have seen, limitations and exceptions to this,
such as the fact that at the organizational level, PayPal was fairly intoler-
ant of various dimensions of difference internally, though those dimen-
sions were themselves offbeat as an early employee described it, Were
all a little weird (OBrien, 2007, p. 106).
Likewise, consider the application of appreciative intelligence to tech-
nological problems by the young techies. This was a culture of making
things work with whatever materials might be available in the garage
and through social networks (Saxenian, 2000; Audia and Rider, 2005).
It also shows a willingness to fail and to persevere in the face of failures.
YouTube and Yelp founders learned a valuable lesson from PayPal:
The first idea isnt always the best. Yelp was a convoluted e-mail referral
service before becoming a top review site. YouTube started as a video
dating play (OBrien, 2007, p. 106).
Googles Marissa Ann Mayer says that Googles innovation relies on
its fearlessness she launches products early and often without fear
of failure (Elgin, 2005). Referring to Apple Computer and Madonna,
Mayer says Nobody remembers the Sex Book or the Newton. Consumers
remember your average over time. That philosophy frees you from fear
(ibid., p. 90).
A founder of PayPal had so much confidence in his own actions that he
founded a financial transactions company even though when he started,
Peter Thiel didnt know what a chargeback was.... Thats one of the fun-
damental things of any credit card payment system. Chargebacks almost
killed the company (OBrien, 2007, p. 106). At PayPal, the executive team
made up for nonmastery of details with unwavering vision, which inspired
the troops (ibid., p. 106).
Moreover, a can-do approach of getting in and trying things and learn-
ing on the fly inspired other employees to found their own businesses. For
example, Chad Hurley, CEO of YouTube remembers his PayPal days as
an education in business. When he arrived in California with a degree
in art from Indiana University of Pennsylvania, building a successful
company seemed like something other people did. You never think it
could happen to you, Hurley says: But seeing Peter and Max and the
guys come up with ideas and seeing how to make things work gave me a
lot of insight. You may not have a business degree, but you see how to put
the process into effect (ibid., p.106).
We also should not underestimate the fact that appreciative intel-
ligence was frequently applied not individualistically to advance ones
own personal gain at the expense of others, but rather collectively,
holistically, and in a more communitarian way to appreciating, sup-
porting and valuing the success of others. Rather than a scarcity-based
zero-sum-game mindset, Silicon Valley was frequently characterized by a
mindset of sharing, abundance and mutual encouragement and support.
Founders of one firm would support former employees in founding their
own firms with capital, knowledge-sharing and other resources. This
happened within myriad social networks, from the immigrant asso-
ciations documented by Saxenian (2000), to the organizational support
in which employees would be supported by their current and former
employers to start companies (Audia and Rider, 2005; OBrien, 2007).
TRANSFERABLE LESSONS
whether or how they would translate into economic gains (Saxenian, 1999,
2000, 2006; Florida, 2002). Of course, the opportunity to make money
was not squandered, and it may even have been primary for some firms,
but this was relatively rare in the organizational and underlying cultures,
and when it did occur in organizations they failed in disproportionately
high numbers (Baron and Hannan, 2002). In the core incubating culture
spawning the phenomenon of Silicon Valley, money was a secondary
rather than a primary driver for large segments of the population.
Baron and Hannan (2002) showed that the Commitment blueprint
treating employees as persons rather than instrumentally as means was
the most successful within Silicon Valley. This model had the least likeli-
hood of failure, while the autocratic model, with a central exchange rela-
tionship of working for money, had the greatest likelihood of failure. This
blueprint, along with some of the others focused on work for the joy of it,
can be seen as an extension to the organizational level of a counter-culture
that refuses to value money as the end and persons as the means, but
instead values relationships and authentic activities as valuable in their own
terms rather than instrumentally for their monetary potential. Again, this
highlights the fallacy of trying to replicate Silicon Valley in order to create
economic wealth: paradoxically, part of what made Silicon Valley eco-
nomically successful may have been the very fact that many of the original
innovators were not motivated because of money but instead as a result of
intrinsic enjoyment of the technology and the company created around it.
The Silicon Valley experience also challenges other framework assump-
tions of Western management. For example, traditional management
prizes asymptotic levels of efficiency, and yet one of the most important
sources of success in Silicon Valley was the willingness to fail and the
acceptance of failure as part of the process. From a traditional man-
agement perspective, failure is waste, and something that should pro-
gressively be eliminated. Not only has the cult of efficiency in Western
management been eroding personal lives and the ability of employees to
bring their whole selves to bear in the workplace, but also it is, like global
capitals destruction of the regional and local, self-defeating.
Traditional management theories also collude with the hierarchical
structure of salaries based on power and control within the company.
Management ideologies usually rationalize these modes of apportioning
wealth with patently false (and irrelevant: Kohn, 1993) claims that pay is
related to actual contribution to the organization. These claims have now
been dramatically exposed by the US financial meltdown, where CEOs
who ruined their companies continued to reap handsome bonus pay imme-
diately before, during and even after their firms received millions of dollars
in bailout funds from taxpayers or were bought as a distressed sale (as in
the real world? If we are to apply the principle of reverse simulation with
an understanding of the biases of figure and ground phenomena, we will
focus more on the conditions that facilitated the emergence of the learning
than on the content of the learning as such (Thatchenkery et al., 1999).
For example, a certain level of trust had to develop in the small group lab
before a member was willing to give or receive feedback. Therefore, what
should really transfer to the outside world is not a technique of giving
feedback, but the conditions that allow trust to develop in a group. If suf-
ficient trust had been developed between the boss and the employee, the
feedback cited earlier might have been received more positively.
Behara et al. (2008) have applied Weicks notion of reverse simulation in
an organizational context. While designing and implementing two distinct
product delivery processes for a large financial services company in the
Organizational Learning Laboratory at George Mason University, which
one of the authors had founded, a new methodology called Empathic
Knowledge Management was developed. It used the appreciative inquiry
framework (Cooperrider and Srivastva, 1987) and utilized tacit knowledge
of participants for process redesign in a learning laboratory environment.
The focus in the lab was to create conditions that would allow staff from
dispersed units in the organization to come together and share knowledge
so that the process redesign time could be cut by half. Once a prototype
loan processing tool was designed in record time in the Organizational
Learning Lab, the focus was not on applying the new tool in the rest
of the organization, but on figuring out how to create and maintain in
their financial services organization the unique learning climate that had
flourished in the laboratory during their three weeks of stay. In any such
endeavor, it is important to keep in mind not just the similarities but also
the differences between the original and transferred context at every level
from content to process to the being of the facilitators, participants and
organizational actors and cultures. It is also important to remember that,
since human and social systems are far more interdependent and complex
than we will be able to theorize in our (necessarily) simplified views and
models of them, whatever we think we know is but a small and fallible
veneer on what we dont know.
When other regions see value in what Silicon Valley has done, they
need to look not just at the superficial level of success factors or salient
characteristics, but also at:
Perhaps a region that has not become highly commercialized can recog-
nize, celebrate and support the contexts and cultures in which people can
continue to express themselves and be motivated intrinsically as opposed
to extrinsically (e.g., in which people engage in activities fundamentally
as ends in themselves vs. instrumentally as means to other ends such as
making money; Kohn, 1993). Perhaps a citys residents can reframe and
examine their unique strengths and regional advantages to come up with
new services and innovation that will further support the community?
What does it look like to reframe the fear of failure into embracing failures
as learning opportunities?
Appreciative intelligence and the embeddedness of human social and
cultural realities suggest that the real message and transferable lessons
from Silicon Valley are the following:
1. Each locale should appreciate, value and embrace its own unique
character and beauty, and be fully, thoroughly and elegantly itself
rather than trying to imitate models plucked from a distant and
foreign ecology.
2. That when communities see aspects of life they appreciate in other
places, they should be circumspect and reflective in why and how
they may adopt these elements and what cascading unintended effects
adoption might have.
3. Value is not unitary but has many and diverse forms, and thus
economic riches are not necessarily even the ideal or goal for a
community to seek.
While the venture capitalist John Doerr may announce to the world
that Silicon Valley represents the largest legal creation of wealth in the
history of the planet, we must not forget that it is only one form of wealth
that he is noticing. If we extend our conception of wealth, making it more
diversified and pluralistic to include other forms of wealth such as spiritual
and relational wealth, there are a number of Silicon Valleys in the world
that we are not celebrating. It is only in our modern Western discourse
that the creation of material and technical wealth has become an obses-
sion held out for all humanity to replicate, and that the pursuit of its
accumulation and concomitant concentration in certain regions appears
to be a goal worth replicating (Sachs, 1992; Escobar, 1995). Moreover, a
The Valley is the distilled essence of entrepreneurial energy. Its ethos is simple:
If its not new, its not cool; if its not cool, its not worth doing. If you dont
own shares, youre getting screwed. If youve been in the same job for more
than two years, your career is over. If you havent been through an IPO, youre
a virgin. This is where a $2 million house is a teardown. This is where a Porsche
is just one more compact car and sushis just another fast food. Never has so
much wealth been created in so little time by so few people. If the Valleys
residents pause to think about it for even a nanosecond, they know theyre as
blessed as those who lived in Italy during the Renaissance. Like the Florentines
and Venetians, theyre building a new age an age of virtual presence, of glo-
bally interconnected communities, of frictionless commerce, of instantly acces-
sible knowledge and stunningly seductive media. (Hamel, 1999)
REFERENCES
Acs, Z.J. and Audretsch, D.B. (1987), Innovation, Market Structure, and Firm
Size, The Review of Economics and Statistics, 69(4), 56774.
Adams, S.B. (2005), Stanford and Silicon Valley: Lessons on Becoming a High-
tech Region, California Management Review, 48(1), 2951.
Alchian, A.A. (1950), Uncertainty, Evolution, and Economic Theory, The
Journal of Political Economy, 58(3), 21121.
Alvares, C. (1992), Science, in W. Sachs (ed.), The Development Dictionary, Zed
Books, London, pp. 21932.
Audia, P.G. and Rider, C.I. (2005), A Garage and an Idea: What More Does an
Entrepreneur Need?, California Management Review, 48(1), 628.
Audretsch, D.B. and Stephan, P.E. (1996), CompanyScientist Locational Links:
The Case of Biotechnology, The American Economic Review, 86(3), 64152.
Baron, J.N. and Hannan, M.T. (2002), Organizational Blueprints for Success
in High-tech Start-ups: Lessons from the Stanford Project on Emerging
Companies, California Management Review, 44(3), 836.
Begg, I. (1999), Cities and Competitiveness, Urban Studies, 36(56), 795809.
Behara, R., Thatchenkery, T. and Kenney, C. (2008), Empathic Knowledge
Management: Reverse Simulation Experiments in a Learning Laboratory,
International Journal of Information Technology and Management, 7(3), 283
314.
Beugelsdijk, S. (2003), Culture and Economic Development in Europe, Doctoral
Dissertation, Tilburg University.
Block, N.J. and Dworkin, G. (1976), The IQ Controversy: Critical Readings,
Pantheon Books, New York.
Blumenthal, H. (1955), The Economic Base of the Metropolis, Journal of the
American Institute of Planners, 21(4), 11432.
Burrell, G. and Morgan, G. (1985), Sociological Paradigms and Organizational
Analysis, Ashgate, Aldershot, UK.
W.F. Miller, M.G. Hancock and H.S. Rowen (eds), The Silicon Valley Edge:
A Habitat for Innovation and Entrepreneurship, Stanford University Press,
Stanford, CA.
Saxenian, A. (2006), The New Argonauts: Regional Advantage in a Global Economy,
Harvard University Press, Cambridge, MA.
Shipley, C. (2006), What Makes Silicon Valley So Different? (19 Oct.), Wisconsin
Technology Network, available at: http://wistechnology.com/articles/3416,
accessed 23 May 2010.
Shiva, V. (2000), Stolen Harvest, India Research Press, New Delhi.
Thatchenkery, T. (2006), Organization Development in Asia: Globalization,
Homogenization, and the End of Culture-specific Practices, in B. Jones
and M. Brazzel (eds), The NTL Handbook of Organization Development and
Change: Principles, Practices, and Perspectives, Pfeiffer/Wiley, San Francisco,
pp. 387403.
Thatchenkery, T. and Metzker, C. (2006), Appreciative Intelligence: Seeing the
Mighty Oak in the Acorn, Berrett-Koehler, San Francisco.
Thatchenkery, T., Behara, R. and Kenney, C. (1999), Building Capabilities for
Change through Laboratory Simulations, Developments in Business Simulation
and Experiential Learning, 26, 1446.
Weick, K. (1983), Utilization as Reverse Simulation: Making the World More
Like the Laboratory, in R.H. Kilmann, K.W. Thomas, D.P. Slevin, R. Nath
and S.L. Jerrell (eds), Producing Useful Knowledge for Organizations, Praeger,
New York, pp. 494520.
Weiss, J. and Delbecq, A. (1987), High-technology Cultures and Management:
Silicon Valley and Route 128, Group and Organization Studies, 12(1), 3954.
Wimsatt, W. (1976), Reductionism, Levels Organization, and the Mind-Body
Problem, in G. Globus, G. Maxwell and I. Savodnik (eds), Consciousness and
the Brain: A Scientific and Philosophical Inquiry, Plenum, New York, pp. 20567.
Wimsatt, W. (1980), Reductionistic Research Strategies and Biases in the Units of
Selection Controversy, in T. Nickles (ed.), Scientific Discovery, Reidel, Boston,
pp. 21360.
Wimsatt, W. (1994), The Ontology of Complex Systems: Levels of Organization,
Perspectives, and Causal Thickets, Canadian Journal of Philosophy, 20(Suppl.),
20774.
Wimsatt, W. (2007), Philosophy for Limited Beings: Piecewise Approximations to
Reality, Harvard University Press, Cambridge, MA.
Zinn, H. (2003), A Peoples History of the United States, Harper-Collins, New
York.
INTRODUCTION
111
Strong proactive
leadership
That, in turn, might enhance the capacity and capability of the region to:
1. In the late 1970s, Norton and Rees (1979) and Rees (1979) pro-
posed that technology was a prime driver in regional economic
development.
2. From that time, theorists such as Romer (1986, 1990), Lucas (1985),
Barro (1990), Rebelo (1991), Grossman and Helpman (1991) and
Arthur (1994) sought to explain technical progress as it generates
economic development as an endogenous effect rather than accept-
ing the neoclassical view of long-term growth being due to exogenous
factors.
3. Thomas (1975) and later Erickson (1994), among others, showed
how technological change is related to the competitiveness of regions.
4. Norton and Rees (1979) and Erickson and Leinbach (1979) showed
how the product cycle, when incorporated into a spatial setting, may
impact differentially on regions through three stages, namely an
innovation stage, a growth stage and a standardization stage.
The term new growth theory has been coined to describe the transi-
tions such as those referred to above in the evolution of thinking on
regional economic development over the last three to four decades, with
the increasing emphasis being placed on the role of and the importance
of, endogenous factors. Importantly, in the new growth theory models
that have emerged, allowance is made for both agglomeration effects
(economies of scale and externalities) and market imperfections, with the
price mechanism not necessarily generating an optimal outcome through
efficient allocation of resources. Also, the processes of capital accumula-
tion and free trade do not necessarily lead to convergence between regions,
with positive agglomeration effects concentrating activity in one or a few
regions through self-enforcing effects that attract new investment. Most
importantly, the new growth theory allows for both concentration and
divergence.
As discussed by Stimson, Stough and Roberts (2006), these evolution-
ary developments in regional economic growth and development theory
are welcome for regional economic development analysts and planners
because, among other things, they explicitly introduce a spatial dimen-
sion into economic growth theory, a dimension that was ignored in
neoclassical economic development theory. That evolution is particularly
important as the role of regions in national economies has changed signif-
icantly since the 1970s. This has been a result to some degree of the effects
of globalization and structural change and adjustment. Understanding
those processes of change is crucial for analysing and understanding
differentials in the patterns of regional economic performance and for
formulating and implementing regional economic development planning
strategy. But just as important is the need to carefully address endogen-
ous factors and processes to enhance the capacity and capability of a
region to proceed along a path of achieving sustainable development and
growth.
The authors and their collaborators have been seeking to develop and
operationalize a new model framework (see Figure 7.2) to measure and
investigate the determinants of spatial variations in endogenous regional
growth performance. In particular, existing theories of regional economic
development tend to underplay the significance of leadership and other
institutional factors.
Stimson, Stough and Salazar (2003, 2005) have proposed a model
(L) Leadership
Direct effects
Measure and evaluation change
Indirect effects over time
Benchmark performance (e.g.,
regional shift component in
shift-share analysis)
leadership (L);
institutional factors (I);
entrepreneurship (E);
which are seen as creating both catalysts and vehicles for better utilization
of a regions resource endowments (RE) and its market fit (M). The model
may be specified as follows:
Stimson, Stough and Salazar (2005) argue that RED is positively related
to RE, M, L, I and E, but that there are likely to be lead and lag effects
in the short to intermediate run, and perhaps cyclical effects in the longer
run. Thus:
So far there have been two exploratory attempts to apply this model
framework to analyse spatial differentials in regional endogenous growth
performance and identify factors that may explain those variable patterns
in regional endogenous growth performance: (1) In the first application,
Stimson, Robson and Shyy (2005, 2006, 2009b) have conducted an explor-
atory analysis using the OLS technique to estimate a model of regional
endogenous growth performance across non-metropolitan regions in each
of the five mainland states of Australia. (2) In the second application,
Stough et al. (2007) have conducted an exploratory study of variations
in endogenous regional growth across US Metropolitan Statistical Areas
(MSAs).
In both of these model applications, the dependent variable measure
of regional endogenous growth is the regional or differential shift com-
ponent across all industry sectors derived from a shift-share analysis of
employment change over a specified period of time, standardized by size
of the regional labour force. Both studies employ a standard OLS tech-
nique to estimate the model of regional endogenous growth. This depend-
ent variable is regressed on a set of endogenously developed regional
variables on demography and economy, and, where possible leadership,
institutions and entrepreneurship, that are selected as explanatory vari-
ables. Both static point-in-time values around the base year and dynamic
change-over-time values over the period of time being studied are used to
model these explanatory variables, which are assumed a priori to be the
possible determinants of regional endogenous growth.4 While stepwise
regression is the core of the analytical methodology, OLS is used to esti-
mate the fit at each iteration as variables are stepped out of the initial
or general starting model. In that modelling, consideration is given to
testing for spatial autocorrelation to address spatial proximity/spillover
effects.
In the sections that follow we provide a summary discussion of the
results of these two exploratory model applications.
The purpose of the Stimson, Robson and Shyy (2005, 2006, 2009b)
exploratory model application was to empirically examine and model pat-
terns of regional endogenous employment growth in Local Government
Areas (LGAs) across non-metropolitan areas of the five mainland states
of Australia and to identify which factors might explain the spatial vari-
ations in the level of endogenous growth performance over the decade
1991 to 2001. It was restricted to using secondary data on regional
demographic and economic characteristics derived from the census.
Model Variables
Stimson, Robson and Shyy (2005) mapped and analysed the spatial
patterns of variation in the regional endogenous employment growth
(or decline) performance over the decade 1991 to 2001 of the non-
metropolitan regional LGAs in the five mainland states of Australia on
the REG_SHIFT dependent variable of endogenous growth. It was found
that:
Model Results
The OLS general model run for all five of the mainland states pro-
duced high R-squared values ranging from top values of 0.98 for South
Australia, 0.96 for Victoria and 0.92 for Western Australia, to the slightly
lower values of 0.88 for Queensland and 0.85 for New South Wales. The
general model thus explains much of the variance in dependent variables
on regional endogenous growth measured by the regional shift in employ-
ment change over the decade 19912001 across the non-metropolitan
LGAs. The results from the general model indicated not only some com-
monality but also a degree of variability in magnitude and direction of the
effects of the independent variables in explaining the spatial variation in
levels of endogenous employment growth/decline across non-metropolitan
regional LGAs in the five mainland states analysed.
For the application of the specific model, the adjusted R-squared values
derived are high, explaining as much as 98 per cent of the variance for
South Australia, 96 per cent for Victoria and 92 per cent for Western
Australia, 88 per cent of the variance for Queensland and 86 per cent for
New South Wales.
The test conducted to investigate the degree of spatial dependence
showed that only in South Australia is any significant degree of spatial
dependency apparent with the Morans I statistic being 0.154, which is
well below the maximum value of 1. For the other states the Morans I
statistic varies from 0.012 for Victoria to 0.051 for Western Australia.
Not surprisingly there are differences between the five mainland states
in the number of significant variables and the mix of independent variables
Independent New South Wales Victoria Queensland South Australia Western Australia
125
POP_CH POP_CH + POP_CH + POP_ CH + POP_CH + POP_CH +
LQMAN_91
LQMAN_CH LQMAN_CH+ LQMAN_CH+
LQPBS_91 LQPBS_91 + LQPBS_91 + LQPBS_91 +
LQPBS_CH LQPBS_CH +
LQPER_91 LQPER_91 + LQPER_91 + LQPER_91 + LQPER_91 +
LQPER_CH LQPER_CH +
UNIQUALS_91 UNIQUALS_91 UNIQUALS_91 UNIQUALS_91
UNIQUALS_CH UNIQUALS_CH + UNIQUALQ_CH+ UNIQUALS_CH +
TECHQUALS_91 TECHQUALS_91 TECHQUALS-91 TECHQUALS-91 +
TECHQUALS_CH TECHQUALS_CH + TECHQUALS_CH +
ROUTW_91 ROUTW_91 ROUTW_91
ROUTW_CH ROURW_CH ROUTW_CH + ROUTW_CH +
INPERS_91 INPERS_91 INPERS_91 +
INPERES_CH INPERS_CH +
20/12/2010 15:12
STIMSON PAGINATION (M2469).indd 126
Table 7.2 (continued)
Independent New South Wales Victoria Queensland South Australia Western Australia
variables
SYBBA_91 SYMBA_91 SYMBA_91 +
126
SYMBA_CH SYMBA_CH
D_COAST D_COAST
D_METRO D_METRO D_METRO
20/12/2010 15:12
An exploratory approach to model determinants 127
0.04
Tweed
Residuals
0.03
Snowy River
Muswellbrook
0.02
Cobar Tamworth
Residuals
Figure 7.4 Order of regional LGA residuals for the specific model: New
South Wales
THE US APPLICATION
The purpose of the Stough et al. (2007) exploratory model was to empiri-
cally examine and model the sources of regional endogenous growth
across the US MSAs for the period 19992002. Specifically, the modelling
attempts to uncover or examine which factors assumed to be endogenous
to a region are actually affecting changes in the regional endogenous
employment growth of 245 MSAs between 1999 and 2002.7
Method
The map in Figure 7.5 shows the pattern of MSA scores on the depend-
ent variable measuring endogenous employment growth/decline for the
period 19992002. On the map the MSAs are differentiated according to
whether their score on the dependent variable was weak or strong, positive
or negative, using the cut-off points method of classification. MSAs are
differentiated according to their population size category.
It is evident that MSAs with different levels of population experienced
different levels of regional endogenous growth. Generally MSAs with
small population size experience relatively higher regional endogenous
economic growth than those with larger populations. However, the
variance for the small-size MSAs is much greater than for the larger-sized
MSAs. Specifically, the regional endogenous growth rate of the medium-
sized MSAs (population size between 0.2 million and 1 million) is lower
than that of the MSAs with a smaller population (less than 0.2 million) by
Figure 7.5 Pattern of MSA scores on the dependent variable standardized endogenous employment growth, 19992002,
by population size category
20/12/2010 15:12
An exploratory approach to model determinants 131
9.53 per cent. Further, the rate for larger MSAs (population size between 1
million and 6 million) is even lower than that of the MSAs with population
size less than 0.2 million by 12.52 per cent.
It is interesting that the small-size MSAs dominate in the 20 top-ranking
MSAs with the highest positive scores on the endogenous employment
growth dependent variable, with only four medium-size MSAs with popu-
lations between 200 000 and 1 million being in the top 20 ranked MSAs.
Not one of the large MSAs with populations more than 1 million is in
the top 20 ranked MSAs. In contrast, nine of the large MSAs are in the
bottom 20 ranked MSAs with the greatest negative scores on the endogen-
ous employment growth dependent variable, and only one of the small
MSAs was in that list of 20 bottom-ranked MSAs.
For the application of the general model, the R-squared value for the
final general regression model was 0.57, with an adjusted R-squared value
of 0.55, meaning that about 55 per cent of the variance in the dependent
variable is explained by the final general OLS regression model, which
includes 11 significant explanatory variables. Table 7.3 lists the variables
used in the model, and Table 7.4 gives parameter estimation and statistical
fit information for the model.
From Table 7.4, some underlying factors capable of affecting regional
endogenous economic growth may be specified and explained. A summary
of the results follows.
Stough et al. (2007) show that four sets of variables measuring resource
endowments (POP_CH, BACH_00, DOCT_00), market fit (LQGOV_98,
INPERS_99, INPERS_CH), institutions (LGOVEM_CH) and entre-
preneurship (EM1_4_CH, EM5_9_CH) have differing levels of effect on
regional endogenous growth for the US MSAs over the study period. For
example, the following outcomes were identified:
132
EDUMP_00 Percentage of Population over 25 with a Master Degree or Professional Degree in 2000 CENSUS
DOCT_00 Percentage of Population over 25 with a Doctoral Degree in 2000 CENSUS
20/12/2010 15:12
ROUTW_CH Percentage Change of Total Occupations (all persons) as Routine Production Workers from BLS
1999 to 2002
INPERS_CH Percentage Change of Total Occupations (all persons) as In-person Service Workers from BLS
1999 to 2002
SYMBA_CH Percentage Change of Total Occupations (all persons) as Symbolic Analysts from 1999 to BLS
133
LGOVEM_CH Percentage Change in Local Government Employments between 1997 and 2002 CENSUS
SOCIAL_CAP Index of Social Capital Averaged during five years from 199398 DDB
20/12/2010 15:12
STIMSON PAGINATION (M2469).indd 134
Table 7.3 (continued)
134
employment of an MSA between 1998 and 2002
20/12/2010 15:12
An exploratory approach to model determinants 135
Table 7.4 OLS general model results for the US MSAs model application
the same period is associated with a 4.10 per cent increase in regional
endogenous growth.
1. While the population factor has significant effects on the regional eco-
nomic growth at all three levels of the US MSAs, the direction of that
effect varies. Population change has a greater impact on the regional
endogenous economic growth of medium-sized MSAs than for large-
size MSAs, while small population size negatively affects the regional
endogenous growth of small-size MSAs. This implies agglomeration
economies may be an important element in the generation of endogen-
ous growth, which is consistent with the literature.
2. Educational attainment level affects regional economic growth with
the highest level of education attainment (measured by the percentage
of population over 25 with a Doctoral Degree in an MSA) positively
affecting regional endogenous economic growth and the lower educa-
tion level (measured by the percentage of population over 25 with a
Bachelor Degree in an MSA) negatively affecting regional endogenous
economic growth.
3. Regional entrepreneurship capital affects regional endogenous growth
of MSAs, but is manifested differently depending on the measure of
entrepreneurship capital used. For small MSAs, the stock value of
the entrepreneurship capital measured by the percentage of employ-
ment in firms with 14 employees positively affects regional economic
growth, while the stock value of entrepreneurship capital measured by
the percentage of employment in firms with 1019 employees nega-
tively affects regional economic growth. Moreover, for small MSAs
the change in entrepreneurship capital also affects regional endogen-
ous growth: the percentage change of the employment in firms with
59 employees in the total employment has a positive impact on
regional endogenous growth over 1999 to 2002, while the percent-
age change of employment in firms with 1019 employees in the total
employment of an MSA between 1998 to 2002 has a negative impact
on regional endogenous growth. Change of regional entrepreneur-
ship capital negatively affects regional endogenous growth for both
medium and large MSAs compared with the small MSAs.
CONCLUSION
In this chapter we have outlined the development by the authors and
their collaborators of a new model framework to measure and investigate
the determinants of spatial variations in regional endogenous growth per-
formance, and in particular to identify the roles of leadership, institutional
factors and entrepreneurship. As exploratory applications of the model,
the sources of regional endogenous growth across the non-metropolitan
regions of the five mainland states of Australia over the decade 1991 to
2001 and across the US MSAs between 1999 and 2002 were empirically
modelled and then identified by a sequenced OLS regression and a step-
wise regression approach. By constructing two OLS endogenous growth
regression models with the regional share of the employment change for
a region as the dependent variable and some variables for measuring
regional resource endowments and market fit in the case of both explora-
tory model applications, plus the effects of variables relating to leadership,
institutions and entrepreneurship characteristics of the region in the case
of the latter exploratory model application, as explanatory variables, we
find that some of the regional endogenous factors do play important roles
in affecting the regional-share employment changes.
Of course much remains to be done in the development of these
operational models that seek to identify the roles of endogenous and non-
endogenous factors in explaining spatial variability of levels of regional
growth/decline performance across the spatial units of a space economy.
That includes the need to more effectively and directly address the issue
of spatial autocorrelation and incorporate weightings in the data matrix
to account for spatial spillover/proximity effects. Further, the problem of
directly measuring the mediating variables of entrepreneurship, leader-
ship and institutions leaves much to be desired and begs for significant
improvements in their measurement. It is interesting in the US case that
metropolitan regional size is found to be an important factor in regional
endogenous growth. The significance of this is that it strongly suggests the
importance of agglomeration forces in contemporary regional growth and
thus further contributes to the confirmation of the validity of the new eco-
nomic geography. This in turn casts additional concern on the importance
of the problem of stimulating and inducing economic growth in general
and endogenous economic growth in rural and non-urban areas.
Finally, the modelling approach might be enhanced by adopting a path-
analytic approach where the intervening/mediating effects of the leader-
ship, institutional and entrepreneurship variables on regional endogenous
growth performance are explicitly tested.
NOTES
* The authors wish to acknowledge the financial support of the Australian Research
Council Linkage International Scheme (grant #LX0346785) and the George Mason
University Foundation for the research project Regional Economic Development and
Performance: Roles of Leadership and Institutional Factors in Endogenous Growth
from which this chapter draws. We also acknowledge the contribution to that research by
Maria Salazar, Alistair Robson, Tung-Kai Shyy, Chunpu Song, Scott Jackson, Jiamin
Wang and Heifung Qian. Finally, we have gained considerably from the comments of
two anonymous reviewers.
1. It is important to note that the new model framework does not negate the importance of
exogenous factors and in fact recognizes these traditional sources of economic growth;
rather it emphasizes the largely missed importance of such mediating factors as entrepre-
neurship, leadership and more generally institutions in the development process.
2. It is important to note that OLS regression models need to be modified to incorporate the
effects of spatial spillover/proximity effects and to give consideration to spatial autocor-
relation as regional economic development and growth models use spatial data.
3. The interested reader is invited to review the work by Plummer and Sheppard (2006),
which illustrates both institutional and endogenous theory foundations for many of the
variables suggested here.
4. Some regions may have a high point-in-time value at the start of period of time but have
little change over that period; others may have a low point-in-time value at the start but
have a large change over time. The dynamic change-over-time variables are added to the
analysis to capture the effect of change during a period of time.
5. The decision to model the five mainland states separately does present a problem in that
the interaction effects between regions across state borders is lost. However, in policy
terms there is often a focus on state-level analysis in Australia.
6. Each step involved withdrawing one independent variable from the model. That deleted
variable had the highest probability that its absolute t-value was equal to or greater than
0.05. The model was then regressed, and new estimates of the model were obtained. This
process was repeated until the variable with the highest probability that its absolute
t-value was less than 0.05 was identified.
7. The period between 1999 and 2002 was selected for the availability of data; in particular,
the most recent data for some variables related to endogenous growth are only avail-
able in this period. Also, the US economy went into a recession in 2002, hence research
based on this time period may partly reveal the effect and importance of endogenous
growth factors when the national economy is in relative decline. However, this is likely
minimal because the recession that began in 2002 was by most standards a mild one and
its duration quite time limited.
REFERENCES
Arthur, W.B. (1994), Increasing Returns and Path Dependency in the Economy,
University of Michigan Press, Ann Arbor.
Barro, R.J. (1990), Endogenous Technological Change, Journal of Political
Economy, 98(5), S71S102.
Duranton, G. and Puga, D. (2000), Diversity and Specialization in Cities: Why,
Where and When Does it Matter?, Urban Studies, 37(3), 53355.
Erickson, R.A. (1994), Technology, Industrial Restructuring and Regional
Development, Growth and Change, 25(3), 35379.
Erickson, R.A. and Leinbach, T. (1979), Characteristics of Branch Plants
INTRODUCTION
This chapter is about the role of rents and endogenous entry in economic
growth two issues we feel are not getting the attention they deserve in the
modern literature on endogenous growth. When Romer (1990) presented
his seminal article, his key contribution was to provide a private incen-
tive for doing R&D. A positive knowledge spillover (in the tradition of
Arrow, 1962) from current to future R&D then provided the mechanism
for endogenous growth. But in his model he attributed the full rents of new
entry to a specialized R&D sector that auctions off blueprints for innova-
tions to entrepreneurs. Competition among the latter leaves no rents for
them and their role is reduced to merely commercializing whatever the
R&D sector comes up with.
Aghion and Howitt (1992) on the other hand, put entry central in
their Schumpeterian model of growth. Entrepreneurs appropriate the
full rents to entry in their model, but Aghion and Howitt also have them
do the R&D that generates the knowledge in a tournament. There they
clearly deviate from Schumpeter who sees no role for the entrepreneur
in opportunity creation (Schumpeter [1911] 1934). In that way Aghion
and Howitt leave no rents to motivate R&D by firms who do not aim to
enter the market with new varieties or higher-quality products. In fact
they make a strong point to explain why incumbent firms do not engage
in R&D.
Both models are a huge improvement over earlier growth models. Yet
in the data we observe that the vast bulk of rents from innovation accrue
to the entrepreneur introducing the innovation in the market. And we can
see incumbent firms do the vast bulk of R&D in any market economy.1
It has long been recognized that rents are what motivates an entrepre-
neur to enter. Schumpeter (ibid.) claimed that entry was the key source
142
In order to explore these three issues Acs and Varga (ibid.) examine
three separate and distinct literatures:
The three literatures focus on different aspects but at the same time are
also complements of each other. The new theories of growth endogenize
technological change and as such link knowledge creation and spillovers
to technological change and macroeconomic growth at the aggregate level.
New growth theory has identified knowledge generation (Romer, 1986,
1990; Aghion and Howitt, 1992) through R&D and human capital accu-
mulation (Lucas, 1988) as the ultimate sources of economic growth. The
empirical evidence in support of this view is mounting (Temple, 1999). The
evidence clearly shows that more knowledge generation through R&D
and human capital accumulation is correlated with economic perform-
ance, both across countries and regions and over time, and there is hardly
geographical clusters can be read to support the notion that such knowl-
edge spillovers are generally localized.
The combination of these two adaptations to the original Romer model
implies that regions with strong knowledge creation will experience high
growth only when there are enough entrepreneurial resources around to
get the new knowledge to the market. Our model then generates a testable
hypothesis. Growth should arise in those regions and countries that have
strong knowledge creation and entrepreneurship, whereas knowledge
creation without entrepreneurship and entrepreneurship without knowl-
edge creation generate less growth for the same levels of effort. Our model
predicts that both R&D and entrepreneurial activity will positively affect
growth in a region, but the marginal impact of either depends strongly on
the presence of the other.
Space and time do not permit us to test our hypotheses empirically in
this chapter, but the evidence in the literature is suggestive. The empiri-
cal evidence on the importance of entrepreneurship in explaining pat-
terns of regional growth and development has recently been surveyed in
Versloot and van Praag (2007) and Acs and Varga (2005) present results
for European country data and also find a positive and statistically sig-
nificant interaction effect between R&D and entrepreneurship. Audretsch
and Keilbach (2005) ran a test on German regions showing that indeed a
positive interaction exists at the regional level.
Based on the above-presented empirical results in the literature, however,
we feel it is worthwhile to dwell on the policy implications of our analysis.
There is a case to be made for R&D subsidies, as knowledge generation
has positive externalities, but given the current set of innovation policies in
place in most countries and regions, policy-makers should carefully identify
the bottleneck in the innovation chain from the original knowledge creation
to the final marketing of new products and services before committing more
public resources to the generation of more new knowledge. To assume that
new knowledge will create economic growth automatically and costlessly
is similar to dropping seeds on rock. Only a fraction of the thus developed
valuable knowledge will actually take root. In some regions more knowl-
edge investments may be needed, but we would argue that in most regions
more entrepreneurial resources would improve the effectiveness of existing
knowledge creation and generate more economic growth for the tax dollar.
We now proceed with a general outline of a growth model, where we
zoom in on the innovative process and present the formal economic deci-
sion problems for knowledge creators and commercializers. From that
analysis we can derive our main propositions in an a-spatial context.
After that we discuss the implications of taking into account spatial
considerations in the model.
Consumers
With numbers referring to the arrows in the Figure 8.1, consumers have
two outgoing and two incoming flows:
(1)
Consumers of final Producers of final
good C good C
Consumers in our model receive interest and labor income every period
and spend their income on consumption and the purchase of new bonds.
They maximize a standard log-linear utility function and face a standard
budget constraint:
max :U 5 e0`log [ Ct ] dt
Ct
#
s.t.: Bt 5 wtL* 1 rtBt 2 Ct
where U is the utility index and a dot over the variable signifies a time
derivative. Consumers then choose consumption (and implicitly savings)
in every period following the Ramsey rule, such that a constant fraction
of income is saved when the interest rate is constant and exceeds the rate
of time preference:
#
Ct/Ct 5 rt 2 r
Total consumption equals the sales and production of final goods produ-
cers as we assume the market for final goods clears. The next four arrows
in Figure 8.1 then relate to the behavior of final goods producers, who are
price takers in factor and output markets:
` ` n
bY
LDP 5
wP
21
c (i) a 1b
xD (i) 5 n 1 2a 2b
2 a 1b
(1 2 a 2 b) Y
a c (i)
i50
And the final goods sector will employ R&D workers as long as the wage
is below:5
aYy (A/n) 2g
wR 5
(r 2 w# R /wR 1 gn# /n)
(10) rental costs of the raw capital used in producing intermediate goods
and financed with bonds, rK;
(11) dividends on ownership shares and/or interest on loans equal to the
expected value of rents E0 [ e0 p (i) di ] at entry, to finance start-up
n
The producer will set his price as a mark-up over marginal costs:
r
c (i) 5
12a2b
And as marginal costs are equal for all varieties, all varieties are priced and
consequently used at the same level. Given that producing a new variety
yields positive profits, new entrants have an incentive to commercialize
ideas for new varieties. We assume that this commercialization process is
costly in terms of labor and specify the entry process as:
#
n 5 ALE
The model equilibrates when all flows into and out of the boxes add up
to 0 (which means agents solve their maximization problems given their
constraints and do not leave any resources idle) and prices equilibrate the
supply and demand on the two factor markets. It can be shown that:
NOTES
1. The part that is not done by incumbent firms is done by publicly funded institutions
such as universities or specialized R&D firms as Romer (1990) envisioned them.
2. See Acs (2009) for a detailed and referenced exposition of this framework.
3 There are many definitions of entrepreneurship in the literature. We follow the
Schumpeterian tradition. See, for example, Braunerhjelm (2008) for a recent overview
of this literature.
4. Even if the knowledge stock is allowed to differ among final goods producers it can be
shown that only those that have A=Amax will employ R&D workers and increase their A
such that the firms with lower knowledge stocks will diminish.
5 There is a horizontal demand curve for R&D labor due to the assumed linearity in
R&D labor in the innovation function.
6. See Acs and Sanders (2008) for an application to the role of patent protection in this
context.
7. So far we copied the Romer (1990) structure. In Aghion and Howitt (1992) these
entrants drive out incumbents with a higher-quality version of existing varieties and
entry leads to average-quality improvement not to variety expansion. To keep the
model manageable we follow Romer (1990) here and focus on variety expansion but
we also feel that entrepreneurial entry may introduce improved versions of existing
products.
8. Patent protection is problematic in this model as we assume that the knowledge creator
is not the same agent as the knowledge commercializer. Patents are generally awarded
to the knowledge creator. There is a large literature (Acs, 2008) that stresses the
importance of the individual entrepreneur for the success of new ventures. His unique
combination of cultural background, skills, knowledge, access to finance and other key
resources and not least important, luck, makes it unlikely that any other entrant could
enter the same market and drive down profits to 0 by copying the incumbent.
9. Although in expectation terms the profits flow back to the consumers in Figure 8.1,
the entrepreneur, more often than not, is that specific consumer and the profits are his
expected returns on forgoing labor earnings during the entry stage. We have modeled
this as the entrepreneur issuing stock and bonds to finance his wage costs to reflect the
fact that they take such opportunity costs into account and desire a market-determined
return on their investment.
10. This follows intuitively from the assumption that all final goods-producing firms are
equal, face the same maximization problem, production possibilities, final demand
curve and set of input prices.
11. We discuss the precise set of assumptions we need to make for this result in Acs and
Sanders (2008).
12. And if there are such impediments, that would be a first target for policy. Braunerhjelm
et al. (2010) refer to the knowledge filter when they discuss the physical, cultural, politi-
cal and institutional barriers to such knowledge spillovers.
13. As, we have argued in Acs and Sanders (2008), may well be the effect of stronger patent
and IPR protection.
REFERENCES
INTRODUCTION
160
Asia is often seen as the focus of the driving forces of this period of
globalization. The reason is that the three largest of these enormous newly
industrializing countries, namely China, India and Indonesia, are within
the South and East Asian economic arena. However, the performance of
these different countries in terms of global and regional output and the
role played by these economies is a complex picture. Figure 9.1 helps us to
visualize these differences. In order to understand the relative scale effects
of the Asian economies, we also include the two Australasian economies
of Australia and New Zealand as comparator cases.
If we consider the growth performance of these Asian countries during
the current phase of globalization, as we see from Figure 9.1 it is clear that
China and India are currently the two fastest-growing large economies
in the world, with 2005 growth rates of 10.2 per cent and 9.2 per cent
respectively. Chinas gross national income in 2005 was US$2269.7 billion,
which ranked it as the worlds fifth largest economy in 2005, and almost
identical in size to the UK economy; Indias economy in 2005 was some
US$804.4bn, and ranked as the tenth largest economy in the world, while
Indonesia, with an economy of some US$282.2bn in 2005 is ranked at
23rd. Yet scale and affluence are different things. If we consider per capita
income rather than the gross income, China, with a per capita income of
US$1740 is ranked 128th in the world, Indonesia with a per capita income
of US$1280 is ranked 139th, while India with a per capita income of
US$730 is ranked 158th in the world. As such, the potential for growth in
each of these countries is still enormous.
Economic growth in each of each of these countries has been associ-
ated with deregulation and liberalization, and the massive internal real-
location of production factors between sectors and regions. However, the
link between these internal growth processes and the broader processes
of globalization that are taking place is via the enormous inflows of
foreign direct investment (FDI) from the advanced economies into these
countries. Yet, inward FDI exhibits different levels of relative importance
in different host economies. Amongst developing countries, the trans-
nationality index of openness, which indicates the scale of inward mul-
tinational investment in terms of FDI inflows, stocks, value-added and
employment, relative to total GDP, ranks China 32nd, India 36th, and
Indonesia 38th. In general, across all developing economies, the overall
relative trans-nationality openness of countries to FDI tends to be higher
in small countries and lower in the larger economies, although Indonesia is
relatively more closed than its scale might suggest, while China is relatively
more open than its scale would suggest.
FDI will, however, continue to flow in very large quantities from devel-
oped countries into these countries over the foreseeable future. Recent
Figure 9.1 The size, wealth and growth of the Asian and Australasian
economies in 2006
1378 greenfield FDI projects, or 11.6 per cent of the global total. India
accounted for 981 greenfield projects, representing 8.3 per cent of the
global total, while Indonesia accounted for 93 projects (UNCTAD, 2007).
In order to give a sense of the relative scale of these numbers, in the same
year, as we see in Figure 9.2, the number of greenfield inward FDI projects
in the US was 723, UK 669, France 582. Asian economies, and in particular
China and India, are by far the most important locations in the world for
greenfield FDI projects (McCann, 2009). Yet as we see in Figure 9.2, apart
from China, the order of magnitude of these greenfield project numbers is
still roughly comparable to many of the large advanced economies. China
now has by far the largest number of domestically located multinational
foreign establishments with 42 753 foreign affiliates in 2004 with some 24
million employees. These affiliate establishments are heavily concentrated
in manufacturing. The 24 million employed in China in foreign affiliates
represents one-third of the global total workforce currently employed in
foreign affiliates. This number has increased fivefold from less than five
million in 1991, a number that is equivalent to the current total level of
domestic US employment in foreign affiliates (McCann, 2009).
terms of total R&D expenditure and seventh in the world in terms of busi-
ness R&D expenditure (ibid.).
Amongst developing regions, it is the countries in South and East Asia
that are the major locations for multinational R&D investment. During
200204, of the 1773 inward FDI projects involving an R&D component,
1095 (62 per cent) were undertaken in developing or transition economies,
of which 861 projects (49 per cent) were undertaken in developing Asia
alone. In the case of developing Asia, the share of R&D accounted for by
the foreign affiliates of US multinational firms increased from 3 per cent
of their total foreign located R&D in 1994 to 10 per cent in 2002. Similar
trends are also observable for multinational firms from other developed
economies that are locating R&D-related investments in Asia (ibid.).
Within developing Asia itself, it is China in particular that dominates
inflows of multinational R&D investment, and the impact on China of
these inflows has also been the most marked. Between 1998 and 2002, the
share of total domestic business R&D in China accounted for by foreign
affiliates located there increased from 18 per cent to 22 per cent. The R&D
expenditure associated with this R&D-related inward FDI now accounts
for 13.5 per cent of Chinas total domestic public sector plus private sector
R&D expenditure (ibid.). To get a sense of how important these multi-
national R&D investments are to China, we can observe that the 42 000
foreign affiliates located in China are currently employing some 24 million
Chinese, and this still only represents 3.1 per cent of the total employed
workforce in China. As such, the relative importance of multinational
R&D expenditure to Chinas knowledge-related activities is four times
greater than the relative importance of multinational firms to Chinas
overall employment (ibid.). In contrast, in the case of India, the most
recent reliable estimates indicate that in 1999, multinational R&D expend-
iture accounted for just 3.4 per cent of domestic private sector R&D. The
types of R&D undertaken in different countries and regions therefore
appear to vary. In India, over three-quarters of R&D expenditure is on
services, and primarily on software development, whereas in China, most
multinational R&D focuses on adaptive innovations for the Chinese
market, but there is increasing evidence across Asia that innovative R&D
is growing.
It is clear that the scale of Chinas and Indias growth in both domestic
R&D and also inward FDI-related R&D growth is very significant, and
in both cases these are very much greater than the equivalent growth rates
for Indonesia. However, from Figure 9.3 we can get a sense of the relative
global and regional contribution of Chinas knowledge sectors by con-
sidering other indicators of innovation such as patents and trademarks.
From the perspective of such knowledge indicators, the relative role
played by China and India is very much less than the output growth and
FDI figures imply.
The total 2002 R&D expenditure of developing countries in South, East
and South East Asia including both China and India is only 14 per cent
of the value for Japan and 6.7 per cent of US R&D expenditure (ibid.).
Similarly, if we consider the number of US patents and trademarks granted
to the residents of particular countries during 200103, we see that China
had 1543, while India had 1022, and Indonesia 108. For comparison, the
respective figures for Taiwan and Korea are 20 414 and 12 195. Similarly,
in terms of US patents and trademarks granted to firms or organizations
of particular countries during 200103 we see that India had 558, China
475 and Indonesia 31. Once again, for comparison purposes the equivalent
figures for Taiwan and Korea are 12 606 and 11 152, respectively (ibid.).
Therefore, although amongst the newly-industrializing countries China
and India are the leading knowledge generators, their relative contribu-
tions are not large in comparison to other advanced Asian economies.
In total, the share of global R&D expenditure accounted for by all the
BRIICS countries combined is only 4 per cent, of which China accounts
for more than half of this level. In contrast, the worlds developed econo-
mies account for 94.7 per cent of global R&D expenditure (ibid.). As
such, it will still take a long time for the knowledge base of the newly-
industrializing countries to catch up with that of the advanced economies.
The enormous transformations that are currently taking place within the
newly-industrializing economies also have a very explicit economic geog-
raphy logic to them that is more or less common across each of the coun-
tries. In each case, national economic growth is inherently about regional
restructuring and the role of agglomeration economies is central to these
growth processes.
China
2000, Chinas GDP per capita increased by ten-fold. However, the initial
impetus for Chinas restructuring and growth came from fairly modest
reforms. China began with the introduction of a rudimentary system of
property rights in order to create incentives and only recently gave con-
stitutional recognition to private property (World Bank, 2005). However,
recognizing the need to access global capital, technology and knowledge
assets via inward multinational investment, China has subsequently also
liberalized many rules regarding services and manufacturing industry
ownership. These changes now allow for greater levels of overseas owner-
ship in many advanced sectors and have been instituted because China
is aiming to attract both a broader range and a higher quality of inward
FDI. In particular, the National Economy and Social Development Plan
2005 emphasized the need to improve the quality of FDI by encouraging it
in high-technology industries, advanced manufacturing, modern services,
agriculture and environmental protection. The plan encourages the estab-
lishment of R&D centres, regional headquarters and bases of advanced
manufacturing. It also welcomed the role of FDI in the reform of state-
owned enterprises (McCann, 2009).
The growth in R&D-related FDI investments in China began in 1993
and reached some 700 projects by 2004 amounting to some $4bn in inward
FDI. Most projects were implemented after Chinas accession to WTO
in December 2001 (ibid.). These R&D investments are mainly focused on
technology-intensive industries such as ICT, automotive and chemicals,
and there is clear economic geography logic to these investments, which
are concentrated in a small number of locations. As seen in Figure 9.4, in
2004, Beijing had 189 foreign-owned R&D centres of which 60 per cent
were in ICT, Shanghai had 140 foreign-owned R&D centres of which 91
are in Pudong, and Guangdong and Jiangsu provinces in the south close
to Hong Kong are home to a combined number of over 100 R&D centres
(ibid.).
The fact that multinational R&D centres were being located in Shanghai,
Beijing, Guangdong and Jiangsu provinces displays a clear logic. These
locations are the core knowledge regions that are growing quickly,
and form the major locations for all types of international investment.
Although they cannot yet be described fully as global cities (McCann,
2008), both Beijing and Shanghai exhibit some world city characteristics.
At the same time, in 2003 the south eastern provinces of Guangdong and
Jiangsu individually accounted for 28 per cent and 19 per cent of FDI,
respectively. The reason that FDI in general, and knowledge-related FDI
in particular, was being located in these cities and regions, is because these
are the major growth regions of China (McCann, 2009).
During the 1970s and 1980s, inequality between provinces and also
between urban and rural areas in China fell consistently (Golley, 2007).
Until the mid-1980s, the growth in per capita productivity and expendi-
ture per capita was higher in rural than in urban areas, which suggested a
slow process of ruralurban convergence (Angang et al., 2005). Between
1978 and 1985, the ratio of per capita disposable income between urban
and rural residents had fallen from 2.57 to 1.85, and ratio of per capita
consumption had fallen to just over 2.1 (ibid.). However, from the mid-
1980s onwards this urbanrural ratio has been reversed. By 1990 the ratio
of both per capita disposable income and consumption had risen to above
2.0. Since the economic reforms started in earnest, as expected on the basis
The next highest ratio areas are the regions of Zhejiang 2.1 and Jiangsu
1.8, which are the regions close to Shanghai, followed by the regions close
to Hong Kong or Guangdong 1.8, and then the coastal regions close to
the dominant cities of Fujian 1.6, Liaoning 1.6 and Shandong 1.4 (Golley,
2007). These ratios imply that the dominant cities currently exhibit GDP
per capita levels that are approximately 13 times that of the lowest regions
(Fujita, 2007).
If the three major city-regions of Shanghai, Beijing and Tianjin are
removed then the increase in regional inequality across China is notice-
ably reduced. On the other hand, however, if we group together all of the
coastal regions including the dominant city-regions, then regional inequal-
ity between the coastal and interior regions of China increases even more
dramatically (Golley, 2007). This demonstrates the role played by particu-
lar city-regions in the dramatic growth of China over the last two decades.
In 2000, the coastal region between Beijing and Hong Kong as a whole
produced 71 per cent of Chinas total industrial output. This enormous
output accounted for more than 60 per cent of output in all but two sectors
and at least 80 per cent of output in close to half of the industry sectors,
including 97 per cent of Chinas cultural, educational and sports outputs
(ibid.). The growth of China is a coastal phenomenon. However, even
within the coastal region of China there is a core region, which consists of
the south east regions adjacent or close to the major cities, and represents
broadly an arc of regions bounded by Shanghai and Hong Kong. These
core region provinces grew by more than the coastal region as a whole in
almost all of the sectors in which the coastal region grew (ibid.).
As expected on the basis of earlier arguments, increasing inter-regional
inequality is now a general phenomenon in China. However, in terms of
economic geography, the escalating growth and wealth of certain regions
is also highly associated with the increasing agglomeration of activities
in these regions. Once again, this is predicted by economic geography
arguments. The core regions of the south east are not only the fastest per
capita growth regions, but also they are the regions of the most rapidly
increasing agglomeration. Golley (2007) calculates that that between 1989
and 2000, 26 out of 28 major manufacturing and industrial sectors have
become more spatially concentrated, as reflected by increasing spatial
Gini coefficients. As such, the general trend towards increasing intra-
national inequality across many countries is clearly very evident in China
(McCann, 2009).
Regional economic restructuring in China has meant that poverty
reduction since the mid-1980s has been most dramatic in the eastern
regions, followed by the central regions, with poverty increasing in the
western regions (Angang et al., 2005). However, this is not just an urban
phenomenon. The ratio of per capita farming incomes in the east and
centre regions relative to the west region have also increased between 1980
and 2000, from 1.27 and 1.05, to 1.92 and 1.30, respectively (ibid.). Part
of the reason is human capital. The areas of highest growth are broadly
the regions with highest rates of literacy (ibid.). In addition, disparities in
Chinese income per capita are also exacerbated by a fiscal tax and transfer
system that significantly benefits urban residents (ibid.). More generally,
however, the competition and wealth effects associated with buoyant
regional growth across a range of local sectors tend to spill over to other
local sectors, and agriculture in such buoyant regions also benefits from
this.
India
two countries. In 1950, Chinas share of global trade was 1 per cent while
that of Indias was 2.2 per cent, whereas by 2002, Chinas share of global
trade had increased to 4.8 per cent while Indias had actually declined to
0.8 per cent (Lardy, 2005). In part, these trade performance differences are
because Chinas rapid growth began slightly earlier than Indias growth,
and also because it has been more dramatic than Indias, particularly in
opening trade. Third, there are also major differences between China and
India in terms of their structure of GDP composition (McCann, 2009).
The structure of GDP composition by industry in India is still rather
different from China; there is a much greater emphasis in India on serv-
ices than China, where manufacturing is still relatively more dominant
(Panagariya, 2005). In 1980 in China, agriculture accounted for 30.1 per
cent of GDP, industry for 48.5 per cent (of which manufacturing alone
accounted for 40.5 per cent) and services for 21.4 per cent of GDP. On
the other hand, for India, agriculture accounted for 38.6 per cent of GDP,
industry for 24.2 per cent (of which manufacturing accounted for only 16.3
per cent) and services for 37.2 per cent. Even though both countries have
undergone enormous changes during the last three decades, the legacy
of these inherited structures still remains. In 2002 in China, agriculture
accounted for 15.9 per cent of GDP, industry for 50.9 per cent (of which
manufacturing alone accounted for 34.5 per cent) and services accounted
for 33.2 per cent of GDP, whereas for India, agriculture accounted for
24.9 per cent of GDP, industry for 26.9 per cent (of which manufacturing
accounted for only 15.8 per cent) and services for 48.2 per cent of 2002
GDP (ibid.).
As a result of its different industrial structure and also its English
language advantages, the growth of FDI in India, and particularly the
growth of off-shoring FDI, has been dominated by a range of service
industries, rather than by manufacturing, which has been the case for
China (McCann, 2009). Yet, many aspects of the dynamic growth indus-
tries of India have similar features to China. First, most of the trade of the
Indian and Chinese economies is still in the form of re-exports of finished
or semi-finished products or services produced by multinational firms that
are based in Europe or the US. Second, many of the key growth centres are
dominated by external links with multinational companies. In the Indian
IT industry, which is dominated by the Bangalore region, two-thirds of all
sales are accounted for by foreign-owned multinational affiliates located
there (Scheve and Slaughter, 2007). Third, as India undergoes continuing
regional economic restructuring, firms located in the regions with large
home markets earn higher profits (Kambhampati and McCann, 2007).
This implies that in terms of economic geography, the large home market
effects associated with agglomeration are driving the internal economic
growth and restructuring within the Indian economy.
Indonesia
rural Indonesia is only 0.2 per cent and more than 64 per cent of rural
villages are entirely without telecommunications (ibid.). Like China,
GDP per capita in Indonesia is also closely related to city size, with the
dominant city of Java exhibiting GDP per capital levels that are approxi-
mately 13 times that of the lowest regions (Fujita, 2007). The opening up
of the country is likely to accentuate these differences, except for the case
where multinationals are engaged in primary sector activities located in
other outlying regions.
These spatial disparities in terms of population and infrastructure are
also manifested in terms of firm investment behaviour. Since the emer-
gence of the Indonesian manufacturing industry from the 1970s onwards,
the role and importance of the major urban areas of Java has increased.
Moreover, as deregulation has continued apace, the relative advantage of
these urban regions has increased even further. Notwithstanding the one-
off shock of the 199798 crisis, over the long term more and more firms
have been investing ever-increasing amounts in the urban centres of Java,
the results of which will tend to be increasing regional inequality within
Indonesia. Yet, why firms do not generally relocate from urban Java to
other parts of Indonesia in order to take advantage of the very large varia-
tions in labour prices that exist across Indonesias regions, and even within
individual islands such as Java, is a moot point (Amiti and Cameron,
2007). Although trade protectionism probably played a role in the initial
concentration of production and investment in these regions, the reasons
why such an extreme spatial concentration of investment within Indonesia
continues appears to be related primarily to the existence of agglomeration
economies.
Indonesian manufacturing exhibits extreme spatial concentration in a
small number of regions within Java (Syamwil and Tanimura, 2000). Over
the last three decades these Javan regions have consistently accounted
for over 85 per cent of total national manufacturing output in large and
medium-sized industries (ibid.), and these geographical patterns of invest-
ment and employment are similarly exhibited for both domestic and
foreign-owned manufacturing firms. As such, increasing international
trade and investment openness appears to be associated with greater
disparities among Indonesias regions, as inward investing firms seek out
the pecuniary advantages of associated agglomeration. This is particu-
larly the case in the apparel and textiles industries, which is the sector in
Indonesia that has exhibited the greatest revealed comparative advantage
since the 1980s (James, 2007). If inward technology spillover effects also
operate in these sectors, these are likely to exacerbate the existing advan-
tages of agglomeration in the dominant urban regions of Java (Amiti and
Cameron, 2007).
On the basis of data from the 1980s and 1990s Amiti and Cameron
(ibid.) estimate the agglomeration advantages that arise from vertical
linkages between firms in Indonesia. Taking account of the location of
input suppliers to estimate cost linkages and the location of customers
to estimate demand linkages, Amiti and Cameron (ibid.) show that the
externalities that arise from demand and cost linkages in Indonesia are
quantitatively very important and highly localized. They find that an
increase in either cost or demand linkages from the 10th to the 90th per-
centile increases wages by more than 20 per cent. As such, firms benefit
greatly from proximity to a large supply of inputs and good market
access, and firms with the best supply or market access can afford to pay
more than 20 per cent higher wages than those with the poorest access
(Amiti and Cameron, ibid.). Moreover, the benefits of these vertical link-
ages are extremely highly localized. Only 10 per cent of the market access
benefit spreads beyond 108km and only 10 per cent of the supply access
benefit spreads beyond 262km (Amiti and Cameron, ibid.). At the same
time, firms located in the major urban areas tend to be more vertically
disintegrated and to exhibit strong local linkages with a variety of back-
ward input supply sources than firms in rural areas (Amiti and Cameron,
ibid.).
There are two crucial aspects to the findings of Amiti and Cameron
(ibid.). First, such extreme economic advantages associated with geo-
graphical localization imply that Indonesian agglomeration effects operate
almost entirely within Java. As agglomeration effects are not only associ-
ated with spatial clustering but also with economies of scale, market access
and infrastructure (Syamwil and Tanimura, 2000) these findings suggest
that Indonesian economic growth will continue to become systemati-
cally more oriented towards the urban areas of Java. For a country that
spans 5500km across 900 islands, the prospects for other regions within
Indonesia would appear to be very much more limited because firms
located in Indonesias outer islands are too far away to benefit from the
agglomeration advantages associated with industries on the main island of
Java (Amiti and Cameron, 2007). The logic of these agglomeration effects
therefore means that central government efforts to promote economic
growth in geographically peripheral and low wage regions face enormous
challenges.
The second crucial aspect of the findings of Amiti and Cameron (ibid.)
is regarding the nature of the agglomeration effects in Java. They find that
labour pooling is quantitatively much less important than the existence
of localized vertical linkages and there also appear to be little or no real
technology spillover effects. Regarding these local spillover effects, most
multinational firms in Indonesia are export-oriented and generally do not
CONCLUSIONS
The growth experience of China, India and Indonesia all share some
common elements. Most notably, the economic growth in each of these
countries is dominated by certain regions, in which agglomeration econo-
mies appear to be crucial. As such, major cities are the focus for economic
growth in all three societies. Both labour and capital are increasingly
mobile and the location of activity is shifting dramatically into the major
urban centres. The result of this is that each society is becoming more
unequal, and the gap between the urban and rural areas is increasing
rapidly. In addition, the gap between the dominant urban areas and other
smaller urban areas is also increasing. On the other hand, there is some-
thing of a difference between the countries in the role played by localized
technology spillovers. The positive effects of these local technology spillo-
vers appears to be more marked in China than in India, and in turn the
effect of these in India is far more marked than in Indonesia, where the
evidence for these is very limited indeed.
REFERENCES
Recent Experience with Reform and Growth, International Monetary Fund and
Palgrave, Basingstoke.
Blalock, G. and Gertler, P.J. (2007), Welfare Gains from Foreign Direct
Investment through Technology Transfer to Local Suppliers, Journal of
International Economics, 74(2), 40221.
Blalock, G. and Veloso, F.M. (2007), Imports, Productivity Growth, and Supply
Chain Learning, World Development, 35(7), 113451.
Fujita, M. (2007), Development of East Asian Regional Economies: A View from
Spatial Economics, in M. Fujita, Regional Integration in East Asia from the
Viewpoint of Spatial Economics, Palgrave, Basingstoke, UK.
Golley, J. (2007), The Dynamics of Chinese Regional Development, Edward Elgar,
Cheltenham, UK and Northampton, MA, USA.
James, W.E. (2007), Comparative Advantage in Thailand and Indonesia and
Potential Free Trade Agreements: Implications for Trade Diversion, in M.A.B.
Siddique (ed.), Regionalism, Trade and Economic Development in the Asia-
Pacific Region, Edward Elgar, Cheltenham, UK and Northampton, MA, USA.
Kambhampati, U.S. and McCann, P. (2007), The Regional Performance and
Characteristics of Indian Manufacturing Industry, Regional Studies, 41(3),
28194.
Kumar, V., Kumar, U. and Persaud, A. (1999), Building Technological Capability
Through Importing Technology: The Case of Indonesian Manufacturing
Industry, Journal of Technology Transfer, 24(1), 8196.
Lardy, N. (2005), Trade Liberalization and its Role in Chinese Economic Growth,
in W. Tseng and D. Cowen (eds), India and Chinas Recent Experience with
Reform and Growth, International Monetary Fund and Palgrave, Basingstoke.
Lee, R.C. and Findlay, C. (2005), Telecommunications Reform in Indonesia:
Achievements and Challenges, Bulletin of Indonesian Economic Studies, 41(3),
34165.
McCann, P. (2008), Globalization and Economic Geography: The World is
Curved, Not Flat, Cambridge Journal of Regions, Economy and Society, 1(3),
35170.
McCann, P. (2009), Globalisation, Multinationals and the BRIICS Countries, in
R. Lattimore and R. Safadi (eds) (2009), Globalisation and Emerging Economies,
OECD, Paris.
Panagariya, A. (2005), India in the 1980s and the 1990s: A Triumph of Reforms,
in W. Tseng and D. Cowen (eds), India and Chinas Recent Experience with
Reform and Growth, International Monetary Fund and Palgrave, Basingstoke.
Scheve, K.V. and Slaughter, M.J. (2007), A New Deal for Globalization, Foreign
Affairs, 3447, JulyAugust.
Sjoholm, F. (1997), Technology Gap, Competition and Spillovers from Foreign
Direct Investment: Evidence from Establishment Data, Working Paper No. 38,
WIJS The European Institute of Japanese Studies, Stockholm.
Syamwil, I.B. and Tanimura, P.H. (2000), The Spatial Distribution of Japanese
Manufacturing Industries in Indonesia, Review of Urban and Regional
Development Studies, 12(2), 12135.
Takii, S. and Ramstetter, E. (2005), Multinational Presence and Labour
Productivity Differentials in Indonesian Manufacturing, Bulletin of Indonesian
Economic Studies, 41(2), 22142.
Timmer, C.P. (2004), The Road to Pro-poor Growth: The Indonesian Experience
in Regional Perspective, Bulletin of Indonesian Economic Studies, 37(1), 177207.
INTRODUCTION
In May 2004 the largest enlargement of the European Union (EU) took
place with ten new member states (NMS), mostly from Eastern and Central
Europe, joining it. They were then followed by Romania and Bulgaria in
January 2007. That expansion of the EU entailed an unprecedentedly high
amplitude of inter-regional disparities (Eurostat, 2007), and was of special
significance for the cohesion policy in the 200713 programming period.
In this chapter we discuss the implications of the EU structural assist-
ance to the NMS on regional disparities, special emphasis being placed on
the capacity of those states to absorb the allocated funds. First we provide
a brief outline of the EUs economic and social cohesion policy. We then
look at the influence of the EU enlargement on that cohesion policy in the
200713 programming period, revealing the close links between cohesion
and the EUs regional policy. Next the regional disparities in the NMS
within European context are discussed in connection with the contribu-
tion of the structural instruments to reducing these disparities. Finally
the absorption capacity of the EU funds in the NMS is examined as a
precondition for achieving this goal. A particular emphasis is placed on
the administrative absorption capacity, considering that the institutional
framework created in the NMS for the administration of the EU funds is
expected to play a decisive role for their successful and complete integra-
tion in the EU structures. That is, the focus is on considering the nature of
institutional issues as either enhancing or detracting endogenous factors.
182
whose GNI is less than 90 per cent of the EU average benefit from the
Cohesion Fund.
In this context it is obvious that the NMS are the main beneficiaries
of the renewed cohesion policy: all of them receive allocations from the
Cohesion Fund, while 51 regions out of 55 NUTS 2 regions in the NMS
are funded under the convergence objective. Hence, the big challenge these
countries have to face in the current financial circumstances of the EU is
whether they will be able to use the large amount of allocated funds. And
further on, the question is whether they will be able to promote adequate
economic policies and economic behaviour so as to generate high rates of
endogenous growth and, thus, ensure an effective use of these funds.
The answer to the first question is usually addressed in terms of the
so-called absorption capacity pertaining to the EU cohesion policy.
That defines the degree to which a country is able to effectively and effi-
ciently spend the financial resources allocated via European Funds. In
other words, it expresses the ability of an EU member state to digest and
consume the funds in order to foster its development and thus to improve
its economic and social performance (NEI, 2002; Horvat, 2004).
The absorption capacity can be addressed from the perspective of the
institutional system created in each member state in order to manage the
funds (the supply side) as well as from the perspective of the beneficiar-
ies of these funds (the demand side). The demand side mainly expresses
the ability of the potential beneficiaries public or private to generate
appropriate and acceptable projects (possible to be financed). The supply
side is determined by three main factors, leading to three components
of the absorption capacity, namely macroeconomic, administrative and
financial absorption capacity:
Source: Compiled by the authors using data available from Eurostat, 2007.
by more than one-quarter, and its physical area by more than one-third.
Now with almost 500 million citizens, the EU generates approximately 31
per cent of the worlds nominal Gross Domestic Product (GDP) in 2007.
However, the combined GDP of all new member states has added only 11
per cent to the GDP of the EU-15, whereas the GDP per capita is 13 per
cent lower than before enlargement (see Table 10.1). Compared with the
EU average, the GDP per capita is 35 per cent per cent higher in the US
and 15 per cent higher in Japan.
If population and GDP per capita are considered, it is noticeable that
the NMS have brought about an important burden for the EU: their
populations represent approximately 21 per cent of the total population of
the EU, whereas in all cases the GDP per capita is below the EU average
(see Table 10.2). The worst situation is recorded by Bulgaria and Romania
with less than 40 per cent of the EU average. By contrast, GDP per capita
was 142.55 per cent in Ireland and 131.91 per cent in the Netherlands
(Luxembourg apart).
The changes of priorities in the new financial exercise for the period
200713 have been determined by changes in the overall background of
the EU under internal and external pressures. The new paradigm of the
200713 cohesion policy, as expressed by the European Commissioner,
Danuta Hubner (2007, p. 1), should be the creation of new opportunities
for the future ... rather than a compensation for the past. In fact this new
paradigm reflects the position of the most member states concerning the syn-
tagma competitiveness cohesion, which no longer represents an antinomy
competitiveness, versus cohesion, but a tandem of interdependent objectives.
Objectives
Table 10.2 Population and GDP per capita in the NMS, 2006a
a
Note: EU-10 plus Romania and Bulgaria.
Source: Compiled by the authors using data available from Eurostat, 2008.
EU level for the accomplishment of the Lisbon Strategy. The three objec-
tives are as follows:
to the rest of the EU, which means another 155 regions, including 61
per cent of the EU-27 population. It accounts for 15.8 per cent of the
funds allocated to the cohesion policy. These regions have relatively
high GDP levels, even if both growth and employment rates remain
weak in many regions. The regional development programmes will
strengthen regional competitiveness by supporting economic and
social innovation, knowledge society, entrepreneurship, protection of
environment and risk prevention.
3. European Territorial Cooperation Objective. This objective aims at
reinforcing cooperation at the cross-border, transnational and inter-
regional levels. It involves 2.44 per cent of funds. The objective is
complementary with the other two objectives, with it being possible
for eligible regions to be funded under both of those previous objec-
tives. The aim of this objective is to promote common solutions for
authorities of different countries in the domain of urban, rural and
coastal development, and to the development of economic relations
and the setting up of small and medium-sized enterprises (SMEs).
Structural Assistance
Table 10.3 Regional GDP per capita in the EU-27 in 2004 (in PPS, EU-
27 = 100)
Source: Compiled by the authors using data available from Eurostat (news release,
23/2007, 19 February 2007).
The Disparities
Source: Compiled by the authors using data available from The Economist Intelligence
Unit.
the top and bottom 25 per cent of regions also increased, from a ratio of
2:1 in EU-15 to a ratio of 3:1 in EU-27, and the average level of GDP per
capita was reduced by almost 12 per cent (2004 data). The regional distri-
bution of wealth among the 268 NUTS 2 regions of the EU-27 shows that
the regional GDP per capita (in PPS) relative to the EU-27 average ranges
from 23.58 per cent in North-East Romania to 302.9 per cent in the UK
capital region of Inner London.
GDP per capita is substantially lower in the NMS, where this indicator
is below 50 per cent of the EU-27 average in most regions (31 out of a total
number of 55 regions). A notable exception is Prague (Czech Republic),
which is the region with the highest GDP per capita in the NMS (157 per
cent of the EU-27 average).
In the NMS, the growth rate was especially high in the three Baltic States,
all having an average annual real GDP growth over 6 per cent (generating
an overall growth of 70 per cent in each country), as well as in Poland,
Slovakia, Hungary, Slovenia and Cyprus. The latest members Bulgaria
and Romania experienced long periods of economic decline during the
Within-country Disparities
Country GDP Per Highest/Lowest Population in Regional Gini Number of Structural and Indicative
Capita in 2004, Regional 2004, Inequality Convergence Cohesion Funds, Allocations
192
Republic
Estonia 12 300 1.4 1/1 615 3 404
Cyprus 19 700 0.7 108 213
Latvia 9 800 2.3 1/1 1 031 4 531
Lithuania 11 000 3.4 1/1 1 379 6 775
Hungary 13 800 2.426 10.1 0.173 6/7 2 837 22 890
Malta 16 400 0.4 1/1 81 840
Poland 11 000 2.184 38.2 0.109 16/16 11 202 66 553
Romania 7 200 2.734 21.6 0.159 8/8 19 213
Slovenia 18 300 2.0 1/1 423 4 101
Slovakia 12 200 3.054 5.2 0.246 3/4 1 544 10 912
Sources: Compiled by the authors using data available from Eurostat (2007).
20/12/2010 15:12
Implications of European Union structural assistance 193
Implications
The new financial perspective for the seven years between 2007 and 2013 is
thus for an EU comprising 27 member states that are displaying increased
economic inequalities. Disparities in the levels of development in the
enlarged EU imply the need for assistance to the least developed regions
and member states by means of an appropriate allocation of Structural
and Cohesion Funds. As previously mentioned, the new round of cohe-
sion policy will be focused on investment in a limited number of priorities
organized around the three main objectives; namely convergence, regional
competitiveness and employment and territorial cooperation.
The Convergence Objective is designed to diminish the amplitude of the
inter-regional disparities, focusing mainly on the least developed regions.
Eligible regions for funding under this objective are the current NUTS 2
regions whose GDP per capita (measured in purchasing power parities) is
below 75 per cent of the average GDP in EU-25 for the period 200002. In
EU-27 there are 84 regions in this category belonging to 17 member states.
From these, 51 regions (out of a total of 55 regions in NMS) belong to 11
NMS.
The overall level of allocations available under the Convergence
Objective amounts to 282.8 billion euros, representing 81.5 per cent of the
total budget for the EUs cohesion policy. Within these allocations, 199.3
billion euros are directed to the Convergence Objective aiming to speed up
the convergence of the least developed regions, preponderantly belonging
to the NMS (see Table 10.5, column 8).
The specific level of allocations to each member state is calculated on
the basis of relative regional and national prosperity and the unemploy-
ment rate of the eligible regions. The resulting indicative allocations of the
As demonstrated earlier in this chapter, the NMS will be by far the most
important net beneficiaries of the EU structural assistance funds. The
financial transfers are designed to increase the economic and social cohe-
sion among the member states, mostly via enhancing a faster catching-up
process of the less developed states and regions in terms of income
per capita. This issue is of a particular importance to the NMS, since
Structural Funds are more important when the economy is weak, the
marginal benefit of an efficient use of Structural Funds being higher in less
developed economies (Daianu, 2003).
However, there are experts who question the possibility of effective,
productive absorption of the substantial financial transfers by the former
centralized economies given all their structural, institutional and admin-
istrative problems (Kalman, 2002). Moreover, especially in the academic
debate, some authors doubt about the ability of fiscal transfers to bring
about economic convergence for the current net recipient member states,
or, in general, about whether convergence can be achieved and, even if so,
whether fiscal transfers are best tools for enhancing convergence (Boldrin
and Canova, 2001). In that debate, various convergence concepts (abso-
lute, conditional) have been discussed,5 but given the current options of
the EU cohesion policy, our objective here is to evaluate the capacity of
the NMS to absorb the large amount of allocated funds.
NMS were preparing their accession to the EU. This concern is about
institutional effectiveness.
In 2002 the NEI Rotterdam developed a study commissioned by
the DG-Regio/DG-Enlargement, which proposed a methodology for
evaluating the capacity of the candidate countries to effectively manage
the Structural Funds (NEI, 2002, p. 1). The methodology analysed the
administrative absorption capacity only for the design (of Structural
Funds) phase, considering that it was premature to address the other two
phases, namely: (1) performance, or the extent to which the Structural
Funds have been managed efficiently and effectively; (2) functioning of
Structural Funds.
The design assessment focused on a series of indicators regarding:
management;
programming;
implementation;
evaluation and monitoring;
financial management and control.
Notes:
A = Strong capacity: system ready for the Structural Funds (at least 90%).
B = Sufficient capacity, but weak points should be addressed (7590% from the maximum
score).
C = Capacity not sufficient yet, serious weaknesses must be addressed (5075%).
D = Insufficient capacity, there is no base for administrating the Structural Funds.
Source: Compiled by the authors using the evaluation by Oprescu et al. (2005) and
Horvat (2004).
time before accession, suggesting that the delays could be recovered and
its accession at the beginning of 2007 was still possible (see Table 10.6).
After the 2004 accession wave, further new studies have been carried out,
concentrating on the challenges that NMS had to face in implementing the
structural assistance allocated for 200406. What they reveal is discussed
in what follows.
A study undertaken by McMaster and Bachtler (2005) provides a
comparative analysis concerning how the NMS have accomplished three
essential functions:
development areas and the setting of clear targets and long-term objec-
tives. In most cases short- and medium-term objectives were preferred
over long-term ones and simple and direct interventions were preferred
to complex ones, able to combine several objectives simultaneously. For
example, direct support to enterprises was preferred to setting up services
for businesses, or modernizing the existing transport infrastructure to
developing combined or alternative transport modes.
Regarding the choices between the national and regional dimensions
of development, that is, between interventions at national level meant
to support general development and economic growth and those aiming
at stimulating the endogenous potential of regional and local develop-
ment, Baleanu (2007) shows that they were largely in favour of the
former. While in the EU-15 the regionalization of the Structural Funds
management has been carried on for more than one decade, in almost
all NMS that joined the EU in 2004 the governments have chosen to use
centralized management systems. Regionalization requires the transfer
of many programming and implementation responsibilities regarding
Structural Funds to the regional authorities, which is not to the advan-
tage of the countries with still weak regional and local administration.
For this reason Poland, Czech Republic and Slovakia even gave up the
design-specific regional programmes in the programming period 200406.
Instead they incorporated them into sectoral programmes or into a single
national programme (for example, Joint Operational Programme in the
Czech Republic).
Centralized management can have a negative effect on smooth imple-
mentation of the programmes and, as a consequence, it can diminish
the absorption speed of the funds, as shown by the low absorption rates
recorded. In the autumn of 2006, two years after the programmes had been
launched, an average of only one-third of the funds allocated to NMS-8
was used (see Table 10.7).
According to the n+2 rule, the EU funds could be spent by the end of
2008, based on a system of annual reallocation, so that the final results
regarding the absorption rate will depend to a great extent on the ability of
administration to strengthen its institutional capacity. So far the current
situation shows that after accession the NMS administrations did not
maintain the pace of reform, which resulted in a low ability to manage
public funds. The implementation of advanced human resource manage-
ment systems was in general limited and inconsistent. A survey conducted
by the World Bank (2006) has shown that innovation is still isolated in
public management: the administrative function of general coordination
is at much lower standards compared with advanced countries, being
unable to keep up with the requirements, politicization has reappeared in
Table 10.7 Absorption rate in NMS-8 between May 2004 and September
2006
Source: Compiled by the authors using data available from the European Commission, as
quoted by Baleanu (2007).
administration, new payment and incentive systems for civil servants are
not put into practice.
At the same time, the experience regarding partnership has been dif-
ficult to assess. This is a major principle in the management of Structural
Funds in order to be able to increase the effectiveness of the programmes
and to achieve the commitment of actors involved in various stages of the
programming cycle, as well as to create good practices in administration.
In many cases it has demonstrated that the participation is not authentic,
but mimic and formal, with negative consequences on the partners com-
mitment and on assuming the ownership of projects results (see Baleanu,
2007).
In order to reinforce the administrative reforms in the NMS and in
the old MS whose administrations still do not function at the required
level, the EC has introduced in the programming of Structural Funds
for 200713 a priority concentrating on the modernization of public
service. This is financed by the European Social Fund. The priority aims
to stimulate good governance practices and to strengthen the capacity of
administrations to meet the requirements for planning and implementing
development plans and for increasing the administrative effectiveness of
public service at national, regional and local level.
Supposing that all these measures will lead to a better administrative
capacity, and hence a high rate of absorption of the EU funds, a further
question relates to the demand side and impact of Structural Funds on
the economic and social welfare in the recipient state. In other words, what
are the effects of Structural Funds on economic growth and in achieving
real convergence? Considering the original idea behind regional policy
CONCLUSION
In order to support the proper functioning of the single market and also
to ensure solidarity among its members, the EU cohesion policy has
an overall objective to stimulate the process of reducing the disparities
between states and between regions via the so-called convergence process.
Those disparities have significantly increased since the 2004 and 2007
accession waves. The structural financial assistance associated with the
cohesion policy plays a central role in this process, which has a special
significance to the NMS as the main net beneficiaries of the financial
transfers.
NOTES
1. The financial chapters refer to Common Agriculture Policy, Structural and Cohesion
Funds and financial obligations of a particular country to the EU budget.
2. These are regions where GDP per capita would be below 75 per cent of the EU-15.
3. These regions are distinguished by their low population density and considerable dis-
tance from mainland Europe. There are seven outermost regions: Guadeloupe, French
Guiana, Martinique and Runion (the four French overseas departments), the Canaries
(Spain) and the Azores and Madeira (Portugal).
4. Even if according to the Berlin Summit (Council Regulation (EC) No. 1260/1999, Art.
7, 8) the upper limit for Structural and Cohesion Funds was set up at 4 per cent, for the
200713 period, the European Council (Dec. 2005) decided that the maximum level of
transfers towards individual member states had to be reduced and the upper limit has
been established between 3.71 and 3.2 per cent (and below) depending on the GNI per
head.
5. In this case the debate is around the possibility of market forces let alone to lead to
the convergence of income in the long run.
6. This refers to the effects of huge positive income shocks for the Netherlands in the 1970s.
REFERENCES
Press Releases Rapid (2003), Structural and Cohesion Funds: Acceding Countries
Need to Further Strengthen their Administrative Capacity, available at: http://
europa.eu/rapid/pressReleasesAction.do?reference=IP/03/103i&format=HTM
L&aged=0&language=EN&guiLanguage=en; accessed 26 May 2010. Brussels,
July.
World Bank (2006), EU-8. Administrative Capacity in the New Member States: The
Limits of Innovation, September.
INTRODUCTION
In this chapter we examine the new required policy targets and styles for
assuring regional performance and competitiveness for the European
Union (EU) and provide an empirical analysis of territorial policies for
regional competitiveness, with a focus on investigating territorial capital
and regional growth.
We first address the theoretical basis of the importance of regional
policies. The strongest argument in favour of regional policies lies in the
long-term persistence and even widening of inter-regional disparities. The
starting point in that theoretical reasoning is that regional policies are fun-
damental for the competitiveness of regional economies. Unlike countries,
regions compete on the basis of an absolute advantage principle, and,
whenever non-competitive, regions cannot rely on any automatic mecha-
nism in order to maintain some export specialization. Thus, in the case of
the EU, the fate of regions is mass unemployment and, in the case of insuf-
ficient public income transfers, emigration and possibly desertification.
In this chapter we stress the idea that the factors determining regional
performance do not lie just in each regions internal development capabil-
ity. In fact, among the causes of regional success and failure, one can find,
on the one hand, some pervasive characteristics of the national economy
and, on the other hand, its general performance. This aspect, linked to the
fact that national policies are not space-invariant within each country,
brings us to the conclusion that national policies also can explain regional
performance. That includes interest rate policies, monetary and fiscal poli-
cies driving movements of the exchange rate, and also such policies such
as transportation and TENs (Trans-European Network) policies, policies
204
for R&D and agricultural policies. All such policies bring selective effects
to different typologies of regions due to a wide array of transmission chan-
nels or preferential regional targets, and, therefore will strongly contribute
to the economic performance of regions.
Of course beyond the national component, the crucial component of
regional performance is each regions internal development capability.
These policies are discussed based on a new concept, that of territorial
capital, and that is analysed through a historical development perspective
of policy targets and styles.
Thanks to the existence of a regional growth forecasting model called
MASST built by the authors (Capello, 2007; Capello et al., 2008), we also
present simulations on the following:
The above arguments look strong enough to disprove the optimistic view
about inter-regional convergence processes, mainly expressed by the neo-
classical school, both in its traditional and modern modelling approaches.
huge social and political costs allowing the explosion of regional crises
and the cultural and environmental costs of regional desertification;
the risk of a super-concentration of population in the big urban
areas of lagging regions (aphenomenon that is typical of developing
countries), as a consequence of the crisis of the surrounding areas
and not of the attractiveness of these urban areas, of a push and not
of a pull factor;
the high opportunity cost of adding successful activities in already
successful areas: in a context of full employment, new workers for
new activities are found at the expense of existing activities, while in
weak areas they are drawn from the unemployment reservoir, and
their opportunity cost is close to zero;
the channelling of a wide share of national savings towards the
building and construction industry and real estate speculation in
advanced regions and cities, as a consequence of the migration
processes, subtracting it from more productive uses;
a lower exploitation of the creativity potential of all regional
communities, constrained by the presence of some basic locational
disadvantage (accessibility, services, infrastructure, and so on).
regional success and failure, one can find, on the one hand, some pervasive
characteristics of the national economy and, on the other hand, its general
performance.
The former elements refer to: (1) institutional factors like the perform-
ance of the high functions of the nation state legislative, justice and
government functions to organizational factors like the efficiency of
services of general interest like education, transportation, communication,
health and security services; (2) economic factors like general fiscal pres-
sure, effectiveness of public expenditure, pervasiveness of environmental
regulations and efficiency of contract enforcement procedures, general
price-competitiveness in case of less advanced countries.
The second element linking regional economies to the general perform-
ance of the national economy is represented by the high inter-regional,
within-countries integration, relative to international integration, in terms
of exchange of goods, services and production factors, due to proximity
effects and absence of any kind of institutional or linguistic barrier. All this
is reflected in the empirical evidence in the EU. By and large, the variance
of regional growth rates within countries is lower than the variance of
international growth rates, all regions benefiting from a good short-term
and long-term performance of their national economy.6
This second element looks even more relevant if interpreted in the light
of the goal of reducing inter-regional disequilibria. As illustrated in Figure
11.1, the results of the MASST simulations and foresights on European
regions (see later in the chapter) show that, between now and 2015, dis-
parities in the EU-27 will decrease only thanks to the national growth
component, namely to fast processes on international catching-up (mainly
by new member states), with a slight increase in intra-national disparities
(see Figure 11.1a).7
In conclusion, if aggregate, national development represents an impor-
tant part of regional development, then it follows that first class territorial
development policies reside in sound and consistent policies internal to
each country, addressed towards a pervasive effectiveness of the public
administration and provision of public goods and externalities enhancing
the development capability of all local economies.
A further element, linked to the previous one, that is worth an in-
depth inspection is the impact of macroeconomic and structural policies
managed at the national level in determining regional performance. In
this case, going beyond the open question of the general effectiveness of
these policies, it looks sound to expect highly differentiated effects on
the different regions. In fact, interest rate policies, monetary and fiscal
policies driving movements of the exchange rate, but also such policies as
transportation and TENs policies, excellence policies in R&D, agricultural
0.2
0.15
0.1
0.05
0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Total (EU-27) Between countries (EU-27) Within countries (EU-27)
0.2
0.15
0.1
0.05
0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
for example, are naturally biased in favour of stronger regions, where most
of the favourable preconditions for successful competitiveness policies are
present and where the demand for new transportation infrastructure is
higher and guarantees the highest economic return to investment.
Another typology of national policies, concerning the macroeconomic
sphere, is usually considered as neutral in terms of inter-regional dispari-
ties. On the contrary, selective effects are visible also in this case, due to
inter-sectoral reasons. In fact, a rise in interest rates is likely to hit most
regions specialized in manufacturing and building and construction rather
than tertiary regions; a similar effect may be expected as a consequence of
a real revaluation of the currency, generating wider tensions in industries
and regions characterized by a wider degree of openness to international
trade. Furthermore, a revaluation process is likely to hit more, ceteris
paribus, those regions specialized in labour-intensive industries, as it
raises labour costs expressed in international currency (while the cost of
capital will remain unchanged, at the level determined internationally,
especially if the revaluation process is interpreted as the effect of a strong
and potentially fast-growing economy).
The last and more recent trend deserves some inspection. A cognitive
approach is increasingly superseding the traditional functional approach
to show that causeeffect, deterministic relationships should give way to
other kinds of complex, inter-subjective relationships that impinge on the
way economic agents perceive economic reality, are receptive to external
Functional approach
(substantive rationality)
Development Innovation
factors factors
Cognitive approach
(procedural rationality)
stimuli, can react creatively and are able to cooperate and work syner-
getically. Local competitiveness is interpreted as residing in cooperation,
trust and sense of belonging rather than in pure availability of capital; in
creativity rather than in the pure presence of skilled labour; in receptivity
to new business ideas and organizational styles more than in the presence
of SMEs per se; in connectivity and relationality more than in pure acces-
sibility; in local identity as well as local efficiency and quality of life (Figure
11.2).
The theoretical elements that support the new methodological approach
may be found in the following:
All the above elements which add to, and do not substitute for, more
traditional, material and functional approaches may be encompassed
and summarized by a concept that, strangely enough, has only recently
made its appearance, and has done so outside a strictly scientific context:
the concept of territorial capital. This was first proposed in a regional
policy context by the OECD in its Territorial Outlook (OECD, 2001),
and it has been recently reiterated by DG Regio of the Commission of the
European Union:
Each Region has a specific territorial capital that is distinct from that of other
areas and generates a higher return for specific kinds of investments than for
others, since these are better suited to the area and use its assets and potential
more effectively. Territorial development policies (policies with a territorial
approach to development) should first and foremost help areas to develop their
territorial capital. (CEC, 2005a, p. 1)
In our view territorial capital may be seen as the set of localized assets
natural, human, artificial, organizational, relational and cognitive
that constitute the competitive potential of a given territory. In this
very large sense it encompasses (Camagni, 2008; Camagni and Capello,
forthcoming):
But also a change in policy styles is needed, residing in the goal of:
(a)
Receptivity
0.418 0.127
0.126 0.001
0.002 0.267
0.268 0.523
0.524 0.713
0.714 0.904
0.905 1.049
1.050 1.203
(b)
Entrepreneurship
0.140 0.244
0.245 0.330
0.331 0.429
0.430 0.535
0.536 0.676
0.677 0.866
0.867 1.157
1.158 2.097
(c)
Creativity
0.276 0.238
0.237 0.196
0.195 0.008
0.009 0.024
0.025 0.056
0.057 0.137
0.138 0.233
0.234 0.424
(d)
areas, like Greece and part of Spain, and France. It is evidently lacking
in eastern countries, in peripheral countries in the north (Scandinavian
countries, UK and Italy). Interestingly enough, capital regions, like
Madrid, Lisbon, Paris, Athens, London, Copenhagen, Oslo, Helsinki,
and important agglomerated regions like the regions in Northern Italy,
Barcelona, Cte dAzur, receive a small contribution from territo-
rial capital to regional growth. We are inclined to say that, as with all
productive factors, territorial capital also shows decreasing marginal
productivity.
Total effect
0.261 0.128
0.129 0.413
0.414 0.655
0.656 0.869
0.870 1.044
1.045 1.236
1.237 1.482
1.483 2.514
(e)
CONCLUSIONS
The objective of this chapter was twofold: on the one hand to present the
rationale for regional competitiveness policies, highlighting the theoreti-
cally based reasons for their existence; and on the other to present a his-
torical development perspective of regional policy targets and styles.
The main messages stemming from the analysis can be summarized as
follows. The persistence of regional disparities justifies the relevance of
NOTES
1. The classical equilibrating process relies on downward flexibility of prices and wages,
which is hampered by the existence of national wage contracts in both private and
public structures and by the homogeneity of import prices (remembering that regions
are very open economies). The second, modern process relies on the devaluation of the
currency, and it is automatically excluded in an inter-regional context.
2. The importance of this last reflection is magnified by the consideration that, with the
creation of exchange rate agreements and large monetary unions, states will increas-
ingly lose control over the external value of their currency, and will increasingly behave
inside the unions like regions inside a nation.
3. Krugman speaks of a general competitiveness obsession, mainly referring to the case
of countries.
4. From a modelling point of view, already 15 years ago it was shown that, if the linearity
assumptions of the traditional neoclassical model were abandoned and some non-
linearities and increasing returns introduced, the same neoclassical model could well
accommodate varied possible outcomes both spatial diffusion and concentration
(Miyao, 1987); and also endogenous growth models, which include cumulative pro-
cesses into a neoclassical production function (Romer, 1986), mainly end up with an
inter-regional divergence process.
5. The Italian experience of the first economic boom, 195764, totally concentrated in
only three areas Milan, Turin and Genoa, the north-western triangle, is telling in this
respect: wage and cost inflation, massive migrations from the south, social and political
tensions, deterioration of environmental quality due to excessive speed in economic
transformation, generated the anticipated end of the expansion phase, an early vanish-
ing of previous competitiveness and 15 years of slow growth in these areas up to the end
of the following decade. History might repeat itself.
6. In the simulation experiment of regional growth in the EU up to 2015, realized through
the MASST model (see later in the chapter) this fact is evident, almost all countries
showing a standard deviation in inter-regional growth rates lower than the interna-
tional standard deviation of growth rates of EU-27 countries. This is particularly true
for the 15 old member countries (with the exception of Ireland and the partial exception
of Spain and Portugal, showing an internal standard deviation similar to the interna-
tional one). On the other hand, all 12 new member countries show a higher variability
of internal regional growth rates, bringing support to expectations la Williamson
about the increasing inter-regional disparities in the first phase of a development or
integration process.
7. This general trend is generated by two differentiated evolutions inside the two blocks
of EU countries. In fact, in the 12 new member countries, inter-regional disparities will
increase, due to a strongly widening intra-national dualism between core and periphery
regions, insufficiently counterbalanced by slowly decreasing international disparities.
On the other hand, a partly different trend is likely to characterize the 15 old member
countries: here too intra-national disparities will increase, though at a much slower pace
than in the 12 new member countries, while international disparities will also increase,
for the first time, though only slightly. Some countries in fact, like Ireland, have already
more than completed their catching-up process, and their persisting miracle will widen
international disparities; other countries, relatively wealthy, like Holland and Italy, are
losing momentum, while some others, less advanced, like Greece or partially Portugal,
are not expected to show convincing performances (see Figure 11.1a and 11.1b).
8. An operational methodology for Territorial Impact Assessment of EU policies, along
the lines suggested by the Commission, enphasizing the three dimensions of territo-
rial cohesion, namely territorial efficiency, quality and identity, was recently devel-
oped with the TEQUILA model Territorial Efficiency, Quality, Identity Layered
Assessment; see Camagni (2006).
9. For a wide explanation of MASST, see Capello (2007); Capello et al. (2008).
10. These simulations were performed as part of the ESPON 3.4.2 Project exploring the
Territorial Impact of EU Economic Policies. For the complete report, see the ESPON
website, www.espon.eu; accessed 27 May 2010.
11. Receptivity is here defined as that part of regional growth that is dependent on the
performance of neighbouring regions, like a sort of growth spillover.
REFERENCES
Andersson, A. (1985), Creativity and Regional Development, Papers of the
Regional Science Association, 56(1), 520.
Arkleton Institute, University of Aberdeen (2006), The Territorial Impact of CAP
and Rural Development Policy, ESPON Project 2.1.3., available at: www.espon.
eu accessed 27 May 2000, Luxembourg.
Armstrong, H.W. (2002), Regional Policy: Reconciling the Convergence and
Evaluation Evidence, in J.R. Cuadrado and M. Parellada (eds), European
Union Regional Policy: Reconciling the Convergence and Evaluation Evidence,
Springer Verlag, Berlin, pp. 23172.
Armstrong, H.W. and Taylor, J. (1993), Regional Economics and Policy, Harvester
Wheatsheaf London.
Bagnasco, A. (1977), Le Tre Italie, Il Mulino, Bologna.
Barro, R.J. and Sala-i-Martin, X. (1991), Convergence Across States and
Regions, Brookings Papers in Economic Activity, 22(1), 10782.
Barro, R.J. and Sala-i-Martin, X. (1992), Convergence, Journal of Political
Economy, 100(2), 22351.
Becattini, G. (1990), The Marshallian Industrial District as a Socio-economic
Notion, in F. Pyke, G. Becattini and W. Sengenberger (eds), Industrial Districts
and Inter-firm Cooperation in Italy, ILO, Geneva.
Borts, G.H. and Stein, J.L. (1964), Economic Growth in a Free Market, Columbia
University Press, New York.
Brusco, S. (1982), The Emilian Model: Productive Decentralisation and Social
Integration, The Cambridge Journal of Economics, 6(2), 16784.
Buchanan, J.M. and Tullock, G. (eds) (1982), Towards a Theory of the Rent
Seeking Society, A&M Press, Austin.
Camagni, R. (1991), Technological Change, Uncertainty and Innovation
Networks: Towards a Dynamic Theory of Economic Space, in R. Camagni
(1991), Innovation Networks: Spatial Perspectives, Belhaven-Pinter, London.
Camagni, R. (1992), Development Scenarios and Policy Guidelines for the
Lagging Regions in the 1990s, Regional Studies, 26(4), 36174.
Camagni, R. (1995), The Concept of Innovative Milieu and its Relevance for
Public Policies in European Lagging Regions, Papers in Regional Science, 74(4),
31740.
Camagni, R. (1999), The City as a Milieu: Applying GREMIs Approach to
Urban Evolution, Revue dEconomie Rgionale et Urbaine, 3, 591606.
Camagni, R. (2001), Policies for Spatial Development, in OECD (2001),
pp. 14769.
Camagni, R. (2002), On the Concept of Territorial Competitiveness: Sound or
Misleading?, Urban Studies, 39(13), 2395412.
Camagni, R. (2006), Territorial Impact Assessment TIA: A Methodological
INTRODUCTION
237
2.0
1.5
1.0
Change in share (p.p:s)
0.5
1 2 3 4 6 10 11 19 20
0.0
5 7 8 9 12 13 14 15 16 17 18
0.5
1.0
2.0
Industry
1 Agriculture/forestry & fishing 11 Financial & insurance services
2 Mining 12 Rental/hiring & real estate services
3 Manufacturing 13 Professional/scientific & technical services
4 Electricity/gas/water & waste services 14 Administrative & support services
5 Construction 15 Public adminstration & safety
6 Wholesale trade 16 Education & training
7 Retail trade 17 Healthcare & social assistance
8 Accommodation & food services 18 Arts & recreation services
9 Transport/postal & warehousing 19 Other services
10 Information media & telecommunications 20 Inadequately described/Not stated
Source: Compiled by the author using data available from the Australian Bureau of
Statistics (2007).
per cent) and Western Australia (up 0.6 of a percentage point to 12 per
cent).
The shifts in production over this period are largely replicated in the
shifts in employment. The states losing national share of employment were
New South Wales (1.6 percentage points), South Australia and Tasmania
(both 0.2 of a percentage point). The states recording an increase in their
share of national employment were Queensland (1.6 percentage points),
Western Australia (0.3 of a percentage point) and Victoria (0.2 of a
percentage point).
Over the decade from 1996 to 2006 there was little change in the rela-
tive dominance of the capital city regions compared with their states as a
Table 12.1 Capital city shares of state employment and population, 2006
Source: Compiled by the author using data available from the Australian Bureau of
Statistics (2007).
242
Accommodation & food services 6 5 6 6 6 6 7
Transport\ postal & warehousing 6 5 5 4 4 5 4
Information media & 4 3 2 2 2 3 2
telecommunications
Financial & insurance services 6 5 4 4 4 5 2
Rental\ hiring & real estate 2 1 2 2 2 2 1
services
Professional\ scientific & technical 8 8 7 6 7 7 4
services
Administrative & support services 3 3 3 3 3 3 2
Public administration & safety 5 5 7 6 6 5 7
Education & training 7 7 8 8 8 7 8
Health care & social assistance 9 9 10 12 10 9 9
20/12/2010 15:12
STIMSON PAGINATION (M2469).indd 243
Arts & recreation services 1 2 1 2 2 2 1
Other services 4 5 5 5 5 5 4
Inadequately described/Not 3 3 3 4 4 3 3
stated
243
Total employment (%) 100 100 100 100 100 100 100
Source: Compiled by the author using data available from the Australian Bureau of Statistics (2007).
20/12/2010 15:12
Table 12.3 Capital city change in industry structure share of employment over the decade 1996 to 2006
244
Accommodation & food services 0.2 0.4 0.1 0.0 0.2 0.1 0.0
Transport\ postal & warehousing 0.2 0.1 0.1 0.0 0.2 0.0 0.0
Information media & 0.5 0.6 0.8 0.5 0.7 0.6 0.5
telecommunications
Financial & insurance services 0.4 0.0 0.2 0.0 0.5 0.0 0.2
Rental\ hiring & real estate 0.2 0.2 0.2 0.0 0.0 0.1 0.1
services
Professional\ scientific & 0.7 0.6 0.1 0.4 0.5 0.5 0.4
technical services
Administrative & support services 0.1 0.6 0.4 0.4 0.1 0.3 0.5
Public administration & safety 0.3 0.2 0.3 1.2 1.0 0.5 0.8
Education & training 0.6 0.6 0.2 0.0 0.1 0.4 0.2
Health care & social assistance 1.0 1.1 1.0 1.2 1.0 1.0 1.3
20/12/2010 15:12
STIMSON PAGINATION (M2469).indd 245
Arts & recreation services 0.0 0.1 0.1 0.2 0.2 0.0 0.0
Other services 0.7 0.9 0.8 0.9 0.9 0.8 0.4
Inadequately described/Not 0.3 0.5 0.5 1.6 1.2 0.6 0.9
stated
245
Total employment 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Note: Figures rounded to nearest whole percentage point (0 refers to less than 0.5).
Source: Compiled by the author using data available from the Australian Bureau of Statistics (2007).
20/12/2010 15:12
246 Endogenous regional development
The regional shift effect provides us with the net impact of factors affect-
ing employment from in the region; or alternatively the endogenous
factors. Hence the technique can provide the new insight that is central to
the concern of this chapter. These factors can be useful to planners and
researchers as they can be influenced more so than factors from outside
the region (national shift effect and industry mix effect that are exogenous
to the region).
TS = NS + IM + RS
NS = Ei gn
where:
IM = Ei gi, n
where:
The RS effect (also known as the differential shift effect) is the residual
between the total shift and the combination of the national shift effect and
the industry mix effect. It can be calculated as either the difference between
those three, or more properly as:
RS = Ei (gi, r gi, n)
where:
issue for large Statistical Divisions, such as the capital city metropolitan
areas, as they are largely self-contained. All data is sourced from the
2006 Census of Population and Housing Time Series Profile (Australian
Bureau of Statistics, 2007). There are particular issues with this data set.
These include the facts that over time, geographies and industry clas-
sifications change. The Australian Bureau of Statistics adjusts (called
concordance) for these changes by particular methods. Another issue is
that people filling out the industry of employment on the census form
may not accurately identify the industry of their employer. An example
the Australian Bureau of Statistics (2006, p. 198) provides is a person who
works for a coal mining company as a driver of the companys coal trucks.
The individuals occupation is truck driver. However, the industry of the
individuals employer is coal mining and not transport.
To provide some measure of standardization, the regional shift effect
throughout this chapter is compared with the change in working age popu-
lation. There are issues with this measure, particularly where the level of
employment self-containment varies between capital cities. Nonetheless,
standardizing the regional shift effect provides a relatively good comparative
tool. This is the first time such a standardization method has been applied to
the results of the Haynes and Dinc method to the knowledge of the author.
Results
Using the shift-share method discussed above, the NS, IM and RS effects
were obtained for the five-capital city regions for the period 1996 to 2006.
The results are given in Table 12.4. Two distinct patterns are evident: In
aggregate the five-capital cities recorded negative RS effects in employment
of 15 668 or 1 job per 100 change in working age population. But most
of this was attributable to Sydney, which recorded a negative RS effect
of 113 034 (or 37 jobs per 100 increase in working age population).
Adelaide recorded a much smaller negative RS effect of 17 193 jobs (or
37 per 100 increase in working age population). In contrast, positive RS
effects were recorded in Brisbane (+66 832 or 29 jobs per 100 increase in
working age population), Melbourne (+22 948 or +7 jobs per 100 increase
in working age population) and Perth (+24 714 or +17 jobs per 100
increase in working age population).
When the decade 1996 to 2006 is segmented into the two inter-census
periods 19962001 and 200106 we get the results shown in Table 12.5.
The following conclusions may be drawn:
250
Sydney 219 197 72 322 080 106 10 126 3 113 034 37
Melbourne 286 066 87 267 520 82 4 412 1 22 948 7
Brisbane 202 382 89 126 212 56 9 366 1 66 832 29
Adelaide 68 374 146 83 925 180 1 638 4 17 193 37
Perth 139 700 96 106 380 73 8,598 6 24 714 17
Australia 1 467 862 94 1 467 941 94
Five-capital cities 915 719 87 906 064 86 25 323 2 15 668 1
Rest of Australia 552 143 107 561 798 109 25 323 5 15 668 3
Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.
20/12/2010 15:12
Table 12.5 Capital city regional shift effects: 19962001, 200106 and 19962006
251
Brisbane 66 832 29 14 133 14 51 436 40
Adelaide 17 193 37 8 604 39 7 945 32
Perth 24 714 17 3 665 5 19 419 26
Australia
Five-capital cities 15 668 1 23 333 4 43 643 9
Rest of Australia +15 668 3 23 333 10 43 643 16
Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.
20/12/2010 15:12
252 Endogenous regional development
the manufacturing industries (20 838 jobs or more than the total
negative RS effect);
the construction industry (7468 jobs or 48 per cent of the total
negative RS effect);
the administrative and support services industries (6078 jobs or 39
per cent of total negative RS effects jobs).
The negative RS effects took place in Sydney (22 609 jobs represent-
ing 11 per cent of the 1996 employment level), Melbourne (21 824
jobs representing 9 per cent of the 1996 employment level) and
Source: Compiled by the author using data available from the ABS 2006 Census of
Population and Housing data.
One of the offsetting factors for the negative RS effect in the five-capital
cities in aggregate was the wholesale trade industry (see Table 12.7). The
wholesale trade industry recorded a positive RS effect in employment of
+8629 jobs for the period from 1996 to 2006.
Again the performance varied between the cities:
Source: Compiled by the author using data available from the ABS 2006 Census of
Population and Housing data.
The next highest RS effect for employment impact in the five-capital cities
was in the construction industry (see Table 12.8). On average, the five-
capital cities recorded a negative RS effect of 7468 jobs or 3 per cent of the
1996 level of employment in the industry. This was entirely due to Sydney:
For Sydney, over the period 1996 to 2006, it was equal to +22 395
jobs, which contributed 20 per cent of its total negative RS. It
Source: Compiled by the author using data available from the ABS 2006 Census of
Population and Housing data.
Sydney
Sydney is the capital city with the largest negative RS effect on employment
of 122 993 over the decade 1996 to 2006, which compares with a total
shift of +219 197 jobs (see Table 12.9). A little under two-thirds of the
negative RS effect is attributable to four industries:
Source: Compiled by the author using data available from the ABS 2006 Census of
Population and Housing data.
Despite this there were three industry sectors that recorded positive RS
effects, namely:
Melbourne
wholesale trade industry (+10 960 jobs or 48 per cent of the total RS
effect);
accommodation and food services industries (+6481 jobs or 28 per
cent of total RS effect);
health care and social assistance industries (+5351 jobs or 23 per
cent of total RS effect).
Source: Compiled by the author using data available from the ABS 2006 Census of
Population and Housing data.
Brisbane
Brisbane recorded a strong positive RS effect of +66 805 jobs over the
decade 1996 to 2006 compared with a total shift of +202 825 jobs (see
Table 12.11). Most of this strong positive RS effect was attributable to
these industry sectors:
Source: Compiled by the author using data available from the ABS 2006 Census of
Population and Housing data.
Perth
Perth recorded a positive RS of +24 728 jobs over the decade 1996 to 2006
compared with a total shift of +139 700 jobs (see Table 12.12). This was
mainly attributable to these industry sectors:
Adelaide
In Adelaide over the decade 1996 to 2006 there was a negative RS effect of
17 184 jobs compared with a total shift of +68 374 jobs (see Table 12.13).
Marginally over half of this negative RS effect is attributable to these
industry sectors:
Source: Compiled by the author using data available from the ABS 2006 Census of
Population and Housing data.
Source: Compiled by the author using data available from the ABS 2006 Census of
Population and Housing data.
CONCLUSION
The research reported in this chapter found that negative regional shift
effects in employment in Sydney and Adelaide have been the drivers of
relatively poor employment growth in those cities over the decade 1996 to
2006. In Sydney, Australias largest city, this was particularly so in the final
half of that period after 2001. For Brisbane and Perth, the regional shift
effects were positive over this period. These findings fit within the broader
context of contemporary regional economic trends during this period.
Given its size, the impact of Sydneys relatively poor regional employ-
ment shift effects are important to the employment performance of the
nation as a whole. Most of the negative regional shift effect in Sydney
over the period was attributable to the manufacturing, construction and
retail trade industry sectors. Comparatively, most of the positive regional
shift effect in Brisbane the nations rapidly growing sub-belt metropo-
lis has been attributable to the manufacturing, health care and social
assistance industry and the retail trade industry sectors.
The national economy has experienced significant changes in its indus-
trial structure in recent decades. This is also true for employment in
the five-capital cities (those with populations in excess of 1 million) of
Australia. One of the key determinants of change in employment for
capital cities has been regional shift effects (i.e., due to endogenous
factors). The research reported here has examined the regional shift effects
on employment in the capital city regions for the period from 1996 to
2006, which was the last decade or so of the long boom experienced by
Australia from the early 1990s.
Future research may examine these regional shift effects for different
count methods of employment. The optimal method would be to use
place of employment (i.e., where the jobs are). Additionally, techniques to
measure production at the capital city level may allow future research into
regional shift effects for production in the capital cities.
NOTES
* The research reported in this chapter is part of a wider project analysing endogenous
regional growth performance across Australia funded in a grant to Robert J. Stimson
under the Australian Research Council Discovery Scheme Grant # DP0879819.
1. As defined by Statistical Divisions.
2. By place of enumeration count method.
3. The southern region refers to the Gold Coast Statistical Division, the western region
refers to the West Moreton Statistical Division, and the northern region refers to the
Sunshine Coast Statistical Division.
4. Place of usual residence method is available for most statistical divisions. This data needs
to currently be downloaded one Statistical Division at a time consuming a significant
amount of time to download all the required data. For most Statistical Divisions there is
little difference between the two count methodologies. On balance, I determined that the
place of enumeration count method was the optimal count method to use at this point in
time.
REFERENCES
INTRODUCTION
268
to their impact in the places they are applied. Most evaluations are based
on opinion surveys with central players with no real tangible evidence of
improved or sustainable development relating to investment, production,
employment and the overall standard of living of people in these commu-
nities. The initial idea for this chapter grew out of the idea of being able to
identify how intra-regional endogenous growth is measured, thus influen-
cing the focus on local policy development.
In his keynote address at the annual conference of the Australian and
New Zealand Regional Science Association International (ANZRSA) held
in Manukau in September 2007, Roger Stough referred to the doctoral
research of Mark De Santis (1993) on intra-regional endogenous growth.
De Santis research was on the relationship between leadership, resource
endowment and exogenous and endogenous drivers of regional economic
development. The author became interested in how it might apply to better
understanding of small town development. The focus was on inter-regional
endogenous growth at a scale several times greater than our interest in small
rural towns. Herein lay the complication I was to discover when research-
ing small town development strategies in two rural shires in Victoria. The
question that arose was: Can the factors informing inter-regional success
be applied to small rural towns on an intra-region comparison?
Local government policy-makers, who are the place managers responsi-
ble for ongoing support for these small towns, would like to know whether
these factors also apply. Stimson et al. (2005) have proposed their virtu-
ous circle for sustainable regional development, which also reflects De
Santis findings. Thus, the question is: Do they apply to intra-regional
assessment of small towns in large rural shires?
After a brief discussion of methodology, there is a discussion of some
of the literature relating to community economic development. That is
followed by a review of two rural shires, and several towns in each shire,
which are used as case studies. There is then a discussion of both the rel-
evance of the social capital concept and measurement and the applicability
of the virtuous circle for sustainable regional development model as it
applies to the four case study small towns. The focus in the chapter reflects
a wider research interest in indicators for small town development, which
represents a much smaller spatial scale of analysis than is usual in the
conventional studies on regional development.
Leadership
High Low
Resource High High Moderate
Endowments economic economic
performance performance
Low Moderate Low
economic economic
performance performance
Source: Compiled by the author and derived from De Santis (1993, p. 91).
here was to identify the factors that contribute to small town success
including social and economic success with the aim being to inform
policy-makers as to how best they might work with such places for their
continuing sustainability. Defining success is thus critical. The criteria will
determine what to measure, and begin to value.
Stimson et al. (2005) presented their virtuous circle for sustainable
regional development with ten factors that constitute the basis for iden-
tifying the success of local community approaches to sustainable devel-
opment. They condense them into leadership, institutions and resource
endowments and market fit, and see how they might interact within what
is termed a regional competitiveness performance cube. Those over-
arching factors are not, I believe, inconsistent with the conclusions drawn
by De Santis (1993) and Eastwood (2007). My view is that the ten factors
that constitute the conceptual model provide the basis for finding out the
local development potential of a small town. For example the Stimson
et al. (2005) premise is that strong proactive leadership that helps build
a vision for future development is backed up by strategy, plans and pro-
cesses and facilitates institutional change to enhance regional capacity and
capability, then the local development process is enhanced. That requires
effective institutions and regional infrastructure, and needs processes and
mechanisms for using resource endowments so as to maximize the poten-
tial to tap market conditions. That may lead to sustainable development
(after Stimson et al., 2005).
It was clear in the research on small towns outlined below that many
of these factors are evident in their small town development projects. The
question is: Why dont the local shires use them as part of their ongoing
evaluation?
The literature on social capital is now extensive and central in efforts by
researchers to understand social structures and processes (Putnam, 1993,
2000; Woolcock, 1998; Lin, Cook and Burt, 2001). Putnams argument
that the degree of association or participation by individuals in society is
related to the success of communities is central to the interests reflected
in this chapter. There is a need to know what these processes are in small
rural towns of several hundred people and how it impacts local economic
development: What are the indicators? How do we measure them, and
what does it tell us about what governments can do to facilitate local eco-
nomic success?
Burt (2001) also raises the idea of brokerage across structural holes as
the source of value-added, and closure can realize value buried in holes.
He contrasts processes of external and internal lack of constraint (see
Figure 13.1). The sociometry of these relationships is presented diagram-
matically in Figure 13.2. As will be seen from the discussion below, it is the
D A
High
Disintegrated group of
Maximum
diverse perspectives,
performance
skills, resources
External lack of
constraint Minimum Cohesive group
performance containing only one
Low perspective, skill,
resource
B
Low Internal lack of constraint High
Source: Compiled by the author and derived from Burt (2001, p. 48).
D B
Source: Compiled by the author and derived from Burt (2001, p. 48).
Enns et al.s (2008) findings support Putnams (1993, 2000) view that:
civic participation is related to the development of critical networks that
are crucial in gaining access to embedded resources and developing social
capital (quoted in Enns et al., 2008, p. 37). They also report a limitation to
this social research technique of interest to the research question being
asked in this chapter that: the position generator is not able to explicitly
capture the social processes whereby people utilize their network ties for
social benefit (ibid., p. 36)
Nevertheless, Enns et al.s (2008) research is worthy of greater consid-
eration in investigating the relationship between intra-region, or shire
council, jurisdictions, as it tells us something about the relationship
between, for example, gender, education and the strength of ties people
have both within and outside their communities. Enns et al. support both
Leonard and Onyx (2003) and Woolcock (1998) who also
advocate for balanced networks with no particular emphasis on one type of tie
in a given location ... involvement in groups that span communities leads to
both weak and strong ties located both inside and outside the community, and
thus allows residents to form diverse, stable networks to resource-rich persons.
(Enns et al., 2008, p. 38)
Shire A
The focus of the strategy was to engage citizens in a discussion about facil-
ities and local government service improvement in each town. A deliberate
planning process was set in place, which resulted in specific plans for each
town. The processes used in Shire A were to:
These are outcome, output and process measures, taken in large part from
state government indicators linked to funded programmes. They give no
clear view of the overarching conceptual framework in which they are
located. It is suggested that this is why the attempt to define and use a wide
range of social indicators in local governments in Victoria largely falls on
deaf ears at the local government level as there is no apparent utilization
in the local decision-making.
Shire As community plan development process is the primary source of
information used for these two case study towns in the shire. Each town
was engaged by the shire in a planning process that established a commu-
nity contact team. Other key stakeholders for the planning process were
identified by this group. The shire then developed a community profile for
each town. Those two towns are now discussed.
Town A1
In the case of Town A1 the profile lists population over time, age cohorts,
employment, buildings, birth rate and a half-page profile, which reads like
a tourism brochure. The community future recognizes that the town is on
the edge of the growth area that is the southeast corridor of the Melbourne
metropolitan region. There are and will continue to be a significant pro-
portion of residents who commute elsewhere for employment. Economic
activity reflects the changing land use in small towns around two hours
drive from metropolitan Melbourne. While dairying and beef produc-
tion remain prominent agricultural industries the town now has many
Town A2
This town is similarly located to Town A1 with respect to its proximity
to the metropolitan region, predominant agricultural production around
Shire B
The overarching strategy for this Shire B is found in their award-winning
Community Development Programme.5 From the shires website they
outline the programme as follows:
There is a clear process set out by the shire council for community plan-
ning. This process provides indicators of success and can be measured.
They are largely institutional and should not be overlooked as important
factors contributing to endogenous development.
Town B1
Located within an irrigated agricultural region, Town B1 has been very
successful in leveraging capital works funds from the shire council to
attract additional monies to upgrade infrastructure in the town including
its business centre and street scaping. The town has also established its own
local fuel outlet that operates out of the business centre being staffed by
some 70 volunteers to ensure the locals do not have to drive many kilome-
tres to the next largest town for fuel. Significantly many of these volunteers
worked together to upgrade the fuel outlet to a modern service station.
One of the local town leaders attributes the success to people with
ideas, motivations and leadership. He asserted that its all about getting
excited about the little things. Many of these volunteers also help out at
the Bush Nursing Centre driving patients to other centres for their particu-
lar health needs.
In 1992 the community formed a progress association, and one of their
first projects was community housing for retirees. They now have 12
units completed, with land for a total of 20 units. Another project was to
convert an irrigation drain through the middle of town into an aquatic
centre playground for the community and visitors. They now have an
ornamental lake, an island, rotunda boardwalk and fountain, which make
their town attractive for both residents and visitors. A more recent project
is their living heritage complex. This is an historical centre showcasing the
development of agriculture in the region. Volunteers have together spent
thousands of hours and also raised tens of thousands of dollars for the
centre.
This town is clearly one that has been more successful than most in the
shire in obtaining resources from external sources as well as from amongst
the residents. It is, in fact, held up as a model of small town success and the
stories presented in the media typically show a range of citizens taking an
active role in their community. As the shire organization applied its com-
munity development model to the town it responded readily and was able
to capitalize on the processes put in place.
Town B2
Town B2 is also located within the irrigation schemes off the Murray
River but has been impacted by water restrictions over the last decade.
It contains a major milk manufacturing facility, now with international
owners who control this factory from the states capital city, Melbourne.
Like Town B1, Town B2 has also engaged in a series of planning meetings
to improve local infrastructure and services, ranging from improvements
to the public swimming pool, construction of a multi-purpose stadium
adjacent to the local school, road intersection realignments and signage. It
has also developed several projects independently from the shire-initiated
community planning process. One was to establish a banking service
with the Bendigo Bank as part of a longer-term development to establish
a community bank. Another recent success was the establishment of a
Rural Transaction Centre with support from the Australian government.
The leaders of these initiatives, who were marginal to the council-initiated
process, also developed a major national campaign Sponsor a Cow. The
district development committee, we are referring to here, created this pro-
gramme to give city folk and farmers the opportunity to Build Bridges
between the city and country and for sponsors to learn first hand the extent
of this drought and the impact and affect it is having on the whole popu-
lation (from Sponsor a Cow website). This project has support from the
Victorian Farmers Federation and Dairy Australia. The sponsorship has
a real impact enabling dairy farmers to save some of their breeding stock
and maintain some dignity during this time, which is the worst drought in
living memory. While the programme ran for two years and raised more
than A$100 000 for participating farmers, the stories of families coming up
from Melbourne to visit their cow and meet the dairy farming family tell
of a much more powerful outcome: of urban Australians feeling what it is
to be a dairy farmer in trying environmental and economic times.
Both Town B1 and Town B2 have benefited from the initiatives of the
shire council with its award-winning community development planning
programme, so much so that it has been put on hold because of funding
difficulties across the whole shire programme. This suggests that for most
small towns their expectations are for funding support from elsewhere for
their local projects. Our two case study towns have also shown they are
able to work together beyond the (exogenous) shire support to develop
their own (endogenous) development strategies.
CONCLUSION
The common characteristics of small town development processes in the
two case study shires were the structuring of relationships to address the
provision of basic infrastructure and local leadership through community
planning processes. It is not known which priorities have been agreed by
each council and subsequently developed. But what was learned through
the studies enquiring into four small towns across two shires in rural
Victoria was that local economic development projects have been initiated
in some towns, but are not as evident in the others. It cannot be denied
that such local initiative does not occur in those other places. It was only in
two where a community-wide approach to such development was found.
Coincidentally it was in the shire that won a national award for its com-
munity development programme a few years earlier. In the other shire, the
community planning process is still relatively new. It will be interesting
to learn if local economic development initiatives spin off from the com-
munity planning processes established by the shire and the communities
themselves attend to local social and economic development.
What intra-regional endogenous development factors can be effectively
evaluated? From the discussion of the social capital literature and a brief
review of intra-regional endogenous development in four small towns
across two rural shires there appear to be several characteristics that are
key to local development:
1. The application of the position generator to these places will tell local
government policy-makers who talks to whom about what and how
often they do this. Quite simply is there a sufficient level of engagement
in the community such that they can have a dialogue, a conversation
about what needs to happen and how they can make it work together?
2. The balance of internal community and external networks needs to
be right, and local leaders can influence these networking opportuni-
ties once they appreciate the importance of striking the right balance,
which will be different for different leaders, at different times working
on different issues.
3. Leadership pervades all of these processes. If appropriate individuals
with the skills and motivation to lead are in sufficient number then
De Santis findings for inter-regional success will also apply for intra-
regional success. But just how many leaders are required is a vexed
question. Of course there is another view that leadership is what is
required, something that many people in a community can display.
NOTES
* The author thanks Ms Barbara Beattie who provided excellent research assistance in
obtaining information from both shires and the four towns for the research on which this
chapter is based.
1. See The Study of Small Towns in Victoria Revisited, available at: http://www.dse.vic.
gov.au/CA256F310024B628/0/E352E96216614FDDCA2572CF0021B941/$File/07_08_
Small_Towns_Revisited.pdf; accessed 28 May 2010.
2. These are referred to as Shire A and Shire B.
3. Although one shire has just signalled to its community leaders that their small towns
strategy is currently on hold due to a lack of funding. This notification came about as
we started to interview key players and the purpose of our research became clear. This
is unfortunate as our intentions are to identify how best to measure intra-regional small
town development success to assist local governments in their future planning for these
places.
4. See The Study of small towns in Victoria Revisited, available at: http://www.dse.vic.
gov.au/CA256F310024B628/0/E352E96216614FDDCA2572CF0021B941/$File/07_08_
Small_Towns_Revisited.pdf; accessed 28 May 2010.
5. From the shires website.
REFERENCES
Lin, N., Cook, K. and Burt, R.S. (eds) (2001), Social Capital: Theory and Research,
Aldine De Gruyter, New York.
OConnor, K., Stimson, R.J. and Daly, M. (2001), Australias Changing Economic
Geography: A Society Dividing, Oxford University Press, Melbourne.
Powers, R. (1991), Far From Home: Life and Loss in Two American Towns,
Doubleday, New York.
Putnam, R.D. (1993), Making Democracy Work: Civic Traditions in Modern Italy,
Princeton University Press, Princeton, NJ.
Putnam, R.D. (2000), Bowling Alone: The Collapse and Revival of American
Community, Simon and Schuster, New York.
Sher, J.P. and Sher, K.R. (1994), Beyond the Conventional Wisdom: Rural
Development as if Australias Rural People and Communities Really Mattered,
Journal of Research in Rural Education, 10(1), 243.
Stimson, R. (2002), Planning for Fewer People, available at www.uq.edu.au/
news/?article=3069, accessed 28 May 2010.
Stimson, R.J., Stough, R.R. and Salazar, M. (2005), Leadership and Institutional
Factors in Endogenous Regional Economic Development, Investigaciones
Regionales, 7, 2352.
Woolcock, M. (1998), Social Capital and Economic Development: Toward
a Theoretical Synthesis and Policy Framework, Theory and Society, 27(2),
151208.
INTRODUCTION
Over the past quarter-century there has been much addition to and
improvement in theoretical constructs of endogenous growth models.
Those improvements include addressing human capital characteristics
and investments (Romer, 1986), the role of firms in regional economies
(Markusen, 1985), innovation diffusion (McCombie, 1982), entrepreneur-
ship (Acs and Armington, 2004), local leadership (Stough et al., 2007) and
many others. Of course, this sampling of contributions does not include
notable works by many researchers, nor does it cover the extensive theo-
retical debates centered on regional convergence/divergence. While the
theoretical debates and theorem proofs are rich, relatively little of this
research has employed empirical techniques to test these theories.
The first, and in many respects foremost, challenge in assessing and
quantitatively modeling endogenous growth at the local level is determin-
ing the best measure of growth to serve as a dependent variable. Unlike
national and state-level analyses, which can rely on well-tested measures
of gross product output, there are few demonstrably valid measures
of regional product available from public sector sources for sub-state
regions. Nascent efforts by the US Bureau of Economic Analysis are
just beginning to offer a measure of regional product for Metropolitan
Statistical Areas, but these data are only available for a few areas/years.
The author is not aware of any public sources for local/regional output
estimates for Australia, the UK or Canada. While there are estimates of
US local area output available from several private vendors, these data
can be cost-prohibitive for research requiring hundreds of cases to support
complex endogenous growth regression equations.
286
As noted above, the traditional measure for growth has been one or more
calculations of gross output. This works well for most national-level
studies and is fairly well supported by gross product measures at the state/
provincial level. However, these data are not readily available at the local
level. Some researchers have used measures of personal income for meas-
uring local-level growth (see, for example, Gkritza et al., 2007). From a
data availability standpoint, employment is the most consistently avail-
able measure of local growth, though some caution must be taken based
on these measures. For example, in the US, the Department of Commerce
offers annual estimates of firm employment by detailed North American
Industry Classification System (NAICS) code. These estimates are based
on place of work. In contrast, the US Bureau of Labor Statistics employ-
ment data are based on place of residence. This potential difference is not
great if one defines the subject area based on labor markets.1 However, the
main concern with using employment data as a measure of endogenous
growth is the effect of non-local economic influences.
One of the most useful and intellectually satisfying contributions to the
literature on the practice of endogenous growth research is the methodol-
ogy proposed by Stimson et al. (2002) using the regional share component
of a shift-share equation to represent endogenous growth essentially the
dependent variable in any number of regression models accounting for
regional development. More recently, Stimson et al. (2006) have adapted
an enhanced shift-share model suggested by Haynes and Dinc (1997).
Starting with Rigby and Andersons efforts to include output and produc-
tivity change to account for ambiguity in the source of employment shifts
in the regional component, Haynes and Dinc (1997) offer an approach for
separating the influences of labor and capital in the productivity change
portion of the regional component.
The intention of using the regional shift component is to remove any
portion of employment change that is not attributable to local character-
istics. In most cases, this approach is likely to work very well. However,
there is a possibility that the regional shift component may not sufficiently
partial out the influence of development incentives that are provided by
non-local government entities. Specifically, the problem is caused when
local economic development is advanced through one-off (non-recurring)
grants or other incentives funded through non-local sources such as state
government. Since the incentives are non-recurring, does the incentive
represent an endogenous or exogenous characteristic?
Source: Compiled by the author using information available from Area Development
(www.areadevelopment.com).
One reason that states and local government have increased their use of
incentives has been effective convergence in state and local tax burdens
driven by interstate competition. According to the Tax Foundation, the
average combined state and local tax burden in the US is 10.10 percent of
per capita income (CNN/Money, 2008). Thirty-five of the 50 states have a
combined average state and local tax burden within 1.0 percent (ibid.).
Based on an ongoing survey conducted by the Council for Community
and Economic Research, 25 of the 50 US states have non-targeted3 grants
as a part of their incentive programs (C2ER, 2008).
Bartik (2005) concludes that lowering taxes has a statistically significant
but small effect on business activity levels. This effect is larger for subur-
ban communities in intra-regional competition for employers. Peters and
Fisher (2004) conducted a meta-review (their terminology) of the litera-
ture, reporting that incentives were found in most studies to have ambigu-
ous or very minor impacts. Gabe and Kraybill (2002) found a negative
relationship between employment change and state incentives in Ohio.
There are, of course, many other studies that could be cited, but the results
are consistently unclear. There is no consistent evidence that incentives
lead to growth, especially employment growth. Why then would there be
an issue with the measurement of endogenous growth?
Simply stated, the magnitude of incentives is tracking upwards and
in many cases has become large enough that dispassionate practitioners
strongly feel that they play a role in site location decisions. Admittedly,
this observation is based largely on casual empiricism, but one can find
compelling evidence to support the notion that incentives can trump other
business location factors.
Alabama
Alabama eventually had to borrow money from its own state employees
pension fund to pay for the promised incentives (ibid.).
Cautioned to a small extent by the near financial disaster of the
Mercedes Benz deal, the state of Alabama and local governments offered a
comparatively paltry US$158 million incentive package to Honda in 1999
or about $105 000 per job created (Anonymous, 1999). The states largess
grew once more in the early part of this decade with a state and local
incentive package for Hyundai totaling about US$253 million, or about
US$117 000 per promised job (Beyerle, 2002).
Texas
In Texas, the state legislature set up funding for the Texas Enterprise Fund
to be used by the governors Office of Economic Development as a deal
closer. These funds are effectively grants to attract or retain businesses,
or to promote expansion in a targeted industry. In Fiscal Years 200407
(the state of Texas operates on a biennium budget), the governors office
granted over US$360 million to 39 companies, of which US$324.1 million
has actually been distributed. Tables 14.3 and 14.4 provide information
on the grants. All grants are tied to job creation and capital investment
criteria. So far about US$300 000 has been recovered through clawbacks
that are included in the grant contracts for failure to create the promised
number of jobs.
Included in the Texas Enterprise Fund recipients is Ruiz Foods, a maker
of frozen Mexican food. Ruiz foods relocated from California to the
cityof Denison located in Grayson County, Texas. This will be the subject
of the case study/simulation discussed later in the chapter. In exchange
for locating in Denison and bringing 423 jobs to the local economy, Ruiz
Foods received an Enterprise Fund (EF) grant of US$1.5 million in June
of 2005. The EF grant is in addition to any local incentives such as tax
abatements, grants, permit fee waivers and the like. There have been nine
EF grants of at least US$10 million with two at the US$50 million level
(Texas Instruments and Sematech).
capture the local competitive effect of the sum of all local characteristics.
These local effects are further refined by the Haynes and Dinc (1997) mod-
eling approach. Still, even with these model improvements over traditional
shift-share adjustments, employment change is still supposed to be related
to local characteristics in the regional component. However, the shift-
share methodology has no recognized method for adjusting the regional
component for exogenous incentives.
The presence of a state-level incentive is, in and of itself, not a problem
when assessing local economic development characteristics, as long as the
incentive is available on a recurring basis. Otherwise, the assessment of
any endogenous growth model would be unreliable because the assumed
competitive advantage would not be present in another time period.
Therefore, if the time period covered by the model coincides with a local
firm receiving a large, one-off subsidy, the assessment of regional competi-
tive strengths is potentially rendered invalid. This is especially true if the
incentivized firm represents a substantial change in area employment, such
as the aforementioned Mercedes Benz assembly plant, which could bias
the coefficients of the growth model at unknown magnitudes. To illus-
trate this point I will use the traditional shift-share calculation for food
manufacturing firms (NAICS 311) to assess the economy of the Sherman-
Denison Metropolitan Area (MA) of Texas.
Source: Compiled by the author using data available from the US Department of
Commerce, Census Bureau and County Business Patterns.
Source: Compiled by the author using data available from US Department of Commerce,
Census Bureau, and County Business Patterns.
CONCLUSIONS
only traditional basic sectors. However, the case study showed that the
influence was greatly muted when major industry regional shares are
summed for the subject area.
Several alternatives are offered for addressing the potential problem
including doing nothing, choosing an alternative dependent variable, or
addressing the concern by including appropriate explanatory variables
in the model to partial out the influence of exogenous incentives. It is
suggested that testing alternative dependent variables and new explana-
tory variables may contribute to endogenous growth modeling exercises
beyond addressing the particular concern highlighted in this chapter.
The choice among the offered alternatives, or preferably the develop-
ment of better alternatives, is left to the discretion of the research teams.
However, even if a do nothing alternative is chosen, researchers should
be able to identify within the cases examined in their models instances of
very large exogenous incentives occurring during their study period with
appropriate cautions for the potential of biased coefficients.
There is one additional perspective that could prove important. The
argument presented in this chapter focuses on the potential for assigning
a better assessment of economically competitive endogenous characteris-
tics than is deserved by the subject area. However, another area is equally
affected by the influence of exogenous incentives but in the opposite
direction. The economic area that loses employment because a firm is
lured away by large incentives may indeed have a competitive disadvan-
tage but is it an endogenous weakness? In the US, where the frequency
and magnitude of economic development incentives increasingly distort
rational economic (site location) decision-making, one has to wonder if
endogenous models at a local/regional level can really capture economic
reality.
NOTES
REFERENCES
Acs, Z. and Armington, C. (2004), Employment Growth and Entrepreneurial
Activity in Cities, Discussion Papers on Entrepreneurship, Growth, and Public
Policy, No. 1304, Max Planck Institute for Research into Economic Systems,
Jena, Germany.
Anonymous (1999), Incentive Deal of the Month, The Site Selection Online
Insider, available at: http://www.siteselection.com/ssinsider/incentive/ti9906.
htm; accessed 28 May 2010.
Bartik, T. (2005), Solving the Problems of Economic Development Incentives,
Growth and Change, 36 (2), 13966.
Beyerle, D. (5 April, 2002), State Unveils Hyundai Incentives, Times Daily
(Florence, Alabama), 1B, 2B, available at http://news.google.com/newspapers
?Nid=1842&dat=20020405&id=DWweAAAAIBAJ&sjid=VskEAAAAIBAJ&
pg=5017,478953; accessed 10 June 2010.
Blakely, E. and Bradshaw, T. (2002), Planning Local Economic Development:
Theory and Practice, 3rd edition, Sage, California.
Buchholz, D. (2009), Business Incentives Reform. Case Study: Mercedes Benz:
Deal of the Century, CFED, available at http://www.cfed.org/assets/pdfs/
business-incentives-reform.pdf; accessed 10 June 2010.
Burnstein, M. and Rolnick, A. (1995), Congress Should Stop the Economic War
Among the States, The Region, 9, 320.
C2ER (2008), State Development Incentive Database, see www.c2er.org for
access.
CNN/Money (2008), How Tax-friendly is Your State?, available at http://money.
com/pf/ features/lists/taxesbystate2005/index.html; accessed 28 May 2010.
Gabe, T. and Kraybill, D. (2002), The Effect of State Economic Development
Incentives on Employment Growth of Establishments, Journal of Regional
Science, 42(4), 70330.
Gambale, G. (ed.) (2008), The 22nd Annual Corporate Survey, available at: www.
areadevelopment.com; accessed 28 May 2010.
Gkritza, K., Sinha, K., Labi, S. and Mannering, F. (2007), Influence of Highway
Construction Projects on Economic Development. An Empirical Assessment,
Annals of Reginal Science, 42(3), 54563.
Haynes, K. and Dinc, M. (1997), Productivity Change in Manufacturing Regions:
A Multi-factor/Shift-Share Approach, Growth and Change, 28(2), 20121.
Markusen, A. (1985), Profit Cycles, Oligopoly, and Regional Development, MIT
Press, Cambridge, MA.
McCombie, J. (1982), How Important is the Spatial Diffusion of Innovations in
Explaining Regional Growth Rate Disparities?, Urban Studies, 19(4), 37782.
Mielcova, E. and Cemerkova, S. (2008), Political Power of the Czech
Representatives in the European Parliament, Munich Personal RePEc Archive,
paper no. 12258, available at: http://mpra.ub.uni-muenchen.de/12258/; accessed
28 May 2010.
Peters, A. and Fisher, P. (2004), The Failures of Economic Development
Incentives, Journal of the American Planning Association, 70(1), 2738.
Romer, P. (1986), Increasing Returns and Long Run Growth, Journal of Political
Economy, 94(5), 100237.
Stimson, R., Robson, A. and Shyy, T. (2006), Modeling Regional Endogenous
301
in spatial disparities among regions (see also Baumol, 1986; Barro and
Sala-i-Martin, 1992; Bernard and Durlauf, 1996; Boldrin and Canova,
2001). In particular, there is: b-convergence: a negative relationship
between per capita income growth and the level of per capita income in
the initial period (e.g., poor regions grow faster than initially rich regions);
s-convergence: a decline in the dispersion of per capita income between
regions over time.
The convergence hypothesis in neoclassical economics has been widely
accepted and discussed in the literature, but is critically dependent on two
hypotheses (see Cheshire and Carbonaro, 1995; Dewhurst and Mutis-
Gaitan, 1995): (1) diminishing returns to scale in capital should prevail,
which means that output growth will be less than proportional with
respect to capital; (2) technological progress will generate benefits that also
decrease with its accumulation (i.e., diminishing returns).
A wealth of studies has been carried out to estimate the degree of
b-convergence and s-convergence (see, e.g., Barro and Sala-i-Martin
1991, 1992). The general findings are that the rate of b-convergence is
in the order of magnitude of 2 per cent annually, while the degree of
s-convergence tends to decline over time, for both US states and European
regions. Clearly there is still an ongoing debate worldwide on the type of
socio-economic convergence, its speed, its multidimensional conceptuali-
zation and its causal significance in the context of regional policy measures
(see, e.g., Fagerberg and Verspagen, 1996; Galor, 1996; Fingleton, 1999).
Important research topics in the current literature appear to be: the role
of knowledge and entrepreneurship; spatial heterogeneity in locational
or socio-cultural conditions; and institutional and physical barriers. An
important new topic in the field is group convergence (or club conver-
gence), which means a convergence of sets of regions towards a more
homogeneous cluster.1
We may conclude that the research field of spatial disparities is still
developing and is prompting fascinating policy issues that deserve in-
depth attention in the years ahead. Besides the concern on policy issues,
also in the academic arena too much interest has arisen over the last
decade in spatial development. The degree of scientific consensus as well
as the degree of cross-fertilization of ideas between regional economists
and mainstream economists has led to interesting debates in regional
science circles.
In addition, in a period of globalization (including financial crises)
and of the creation of broad single-currency areas, regions (and also
nations) have to keep an eye on the competitiveness of their production
systems, because no spontaneous or automatic adjustment mechanism
is yet at work to counterbalance a lack (or an insufficient growth rate)
formalized and quantitative way, the latter oriented towards the identi-
fication of all tangible and intangible qualitative elements of the growth
process of regions.
entrepreneurial ability;
local production factors (labour and capital);
the relational skills of local actors.
Dynamic Approaches
Concern for the local environment is not a recent policy issue. Already in
Ancient Rome we find examples of policy measures to mitigate the noise
nuisance of horse-drawn carriages during the night. And several medi-
eval cities in Europe had guidelines on where and when to park carriages
(including priority rules for wealthy citizens). In the welfare economics
literature, such non-market phenomena were called externalities. To
guarantee again an equalization of marginal costs and benefits, a system
of taxes (or subsidies) was foreseen, so that an optimal market equilibrium
could be re-established. These Pigouvian welfare rules were seen as inter-
esting exceptions to an otherwise perfectly working market mechanism.
However, in the age of mass industrialization and large-scale mobility,
such externalities became a dominant phenomenon rather than an excep-
tion. Already in the 1950s and 1960s, the first voices on environmental
decay were heard, often inspired by concerns about declining water quality,
noise nuisance, local health and air pollution. The real breakthrough and
awareness took place in the first wave of environmental consciousness, viz.
after the publication of the First Report to the Club of Rome in 1973. In
that global study, much concern was expressed about environmental pol-
lution, scarcity of natural resources, the threat of the population bomb,
lack of food and the position of the developing countries. The next wave of
increased interest emerged in 1987, with the publication of the Brundtland
Report. This document was greatly concerned about the issue of sustain-
able development, in particular from the viewpoint of the developing
world and from the viewpoint of the next generations. Since then, sus-
tainability has become a fashionable word but, unfortunately, it lacked
an operational definition, so that in practice this concept was uncritically
used for sectoral development, for regional or local developments, or for
development with regard to new generations. However, this concept has
prompted an avalanche of research including applied modelling research
on various elements of local, urban or regional sustainability, from the
perspective that our world calls for a holistic view on future development.
And, finally, we witness the third wave, where the interest in global devel-
opments such as climate change or the rise in sea level is anchored in
local or regional settings. This embeddedness of the global change debate
in local and regional development has prompted new types of research in
the area of regional science (see Batabyal and Nijkamp, 2008). From the
wide range of issues, we now select five important focal points that serve
to illustrate the importance of the recognition of sustainability issues in
regional development.
The welfare of regions and the use of the physical resource base of these
regions are clearly mutually interwoven phenomena. It is thus clear that
regional economic development and sustainability strategies may be seen
as mutually complementary forces that may reinforce each other.
Natural Resources
Smart local resource use may improve local welfare conditions. We may
conclude, therefore, that proper waste management at local or regional
level will create favourable conditions that stimulate balanced local or
regional development.
Environmental Regulation
The relationship between global change and local development has often
been neglected, but is receiving increased attention. Global warming will
certainly have a far-reaching impact on local or regional development
conditions, as illustrated by the following examples.
Changes in our ecosystems with more extremes and outliers will call
for adjusted ecosystems management at local or regional level so that these
systems can be better protected against floods or other nature disasters.
This may call for other types of land use policy.
Another example concerns the interrelationship between local and global
climatic conditions (e.g., in terms of building regulations, or water or energy
supply). This prompts increasingly proactive policies (e.g., in the EU).
Many cities are to be found along the coast or river flood plains. The rise
in sea level calls for new design and security principles in urban planning
in the decades to come. We may clearly draw the conclusion that sustain-
ability policy will call for drastic adjustments of local and regional land
use, environmental and resource policy.
Much quantitative research has been undertaken in the past few decades
to map out the complex policy dynamics of the modern space economy.
Computable General Equilibrium modelling and simulation modelling
have proven to be appropriate tools.
More recently, we have seen the emergence of geographic information
systems (GIS) as an example of operational model applications. A great
many useful applications can be found in the literature. For instance: (1)
Several studies have been published on housing planning and demand,
with respect not only to the quality of the dwellings in a narrow sense,
but also to the neighbourhood and accessibility effects in a broader sense
(Kain and Quigley, 1970). (2) Another example concerns location analysis
in the context of land use and land values.
GIS has been very instrumental in dealing with large spatial data sets,
for example, in evaluating commercial buildings and densities, in making
spatial assessment of land values, or in identifying risk-prone areas
(Geoghegan et al., 1997; Weng, 2002). GIS has indeed become an impor-
tant toolbox for regional science research, but its potential for local or
regional socio-economic development policy is still underutilized. There is
still a long way to go before geo-science modelling will be an integral part
of local or regional development policy.
The interwoven nature of the ecology of our planet at differ-
ent geographic-scale levels with local and regional development thus
prompts many new research endeavours. There is yet one final question:
Has sustainability had an impact on regional development policy, and is
there a link to endogenous growth policy? The answer is in the affirma-
tive. Environmental and resource conditions at local or regional scales are
no longer permanently given factors. They can be influenced and adjusted
by smart and appropriate decisions and investments in both the public and
the private sector. In other words, regional sustainable development from
an endogenous perspective is the result of deliberate choices and actions
of local and regional stakeholders, including public policy-making bodies.
growth is thus a race without a finish; and that also applies to regional
growth theory.
NOTES
1. For further discussion on club convergence, we refer to Baumont et al. (2003); Chatterji
(1992); Chatterji and Dewhurst (1996); Fischer and Stirbock (2006); Islam (2003);
Lpez-Bazo et al. (1999); Quah (1996); Rey and Montouri (1999); Sala-i-Martin (1996).
2. For an extensive review of regional growth theories, see, among others, Johansson et al.
(2001).
3. For the literature on spatial spillovers, see, amongst others, Anselin et al. (1997 and 2000);
Audretsch and Feldman (1996); Aydalot (1986); Feldman (1994); Feldman and Audretsch
(1999); de Groot et al. (2001); Jaffe (1989); Jaffe et al. (1993); Maier and Sedlacek (2005);
on collective learning, see Bellet et al. (1993); Camagni (1991); Capello (1999) and (2001);
Crevoisier and Camagni (2000); Maillat et al. (1993); Rallet (1993); Rallet and Torre
(1995); Ratti et al. (1997); on learning regions, see Lundvall (1992); Lundvall and Johnson
(1994); Maskell and Malmberg (1999); on knowledge-based regions, see Florida (1995);
Malecki (2000); Nijkamp and Stough (2004); Simmie (1997).
REFERENCES
Acs, Z.J., Anselin, L. and Varga, A. (2002), Patents and Innovation Counts as
Measures of Regional Production of New Knowledge, Research Policy, 31(7),
106985.
Aghion, P. and Howitt, P. (1998), Endogenous Growth Theory, MIT Press,
Cambridge, MA.
Anselin, L., Varga, A. and Acs, Z. (1997), Local Geographic Spillovers between
University Research and High Technology Innovations, Journal of Urban
Economics, 42(3), 42248.
Anselin, L., Varga, A. and Acs, Z. (2000), Geographic and Sectoral Characteristics
of Academic Knowledge Externalities, Papers in Regional Science, 79(4), 43543.
Armstrong, H.W. (1995), Convergence among the Regions of the European
Union, 19501990, Papers in Regional Science, 74(2), 14352.
Audretsch, D. and Feldman, M. (1996), R&D Spillovers and the Geography of
Innovation and Production, American Economic Review, 86(3), 63040.
Aydalot, Ph. (ed.) (1986), Milieux Innovateurs en Europe, GREMI, Paris.
Bailly, A. and Coffey, W. (1994), Regional Science in Crisis: A Plea for a More
Open and Relevant Approach, Papers in Regional Science, 73(1), 314.
Barro, R. (1991), Economic Growth in a Cross-section of Countries, Quarterly
Journal of Economics, 106(6), 40743.
Barro, R.J. and Sala-i-Martin, X. (1991), Convergence across States and Regions,
Brookings Papers on Economic Activity, 1, 10782.
Barro, R.J. and Sala-i-Martin, X. (1992), Convergence, Journal of Political
Economics, 100 (2), 22351.
Barro, R. and Sala-i-Martin, X. (1997), Technology Diffusion, Convergence and
Growth, Journal of Economic Growth, 2(1), 127.
Batabyal, A.A. and Nijkamp, P. (2008), Sustainable Development and Regional
325