2001: A SPACE ODYSSEY?
On the supposed loss of spatial concentration of economic activities due to ict
Frank van Oort
National Spatial Planning Agency (RPD), The Hague,
Erasmus University Rotterdam
Lambert van der Laan
Erasmus University Rotterdam
The theme of the ISOCARP conference this year (“Honey, I Shrunk the Space”) concentrates on the enlarged accessibility of information, consumer- and producer goods due to the growing importance of information and communication technology (ict). Complementary to this discussion, the majority of the (popular) spatial-economic literature supposes a redundancy of the necessity for spatial concentration of economic activities. In the visionary film 2001: A Space Odyssey, made by Stanley Kubrick in 1968, the interaction between technology and human activities were central as well. In the movie, the computer HAL takes over control of the spacecraft, that then becomes out of (human) control. After much uncoordinated space-travelling, new forms of life are discovered, but contact with the home base on earth is definitely impossible. Is progress only reachable (or better facilitated) by giving up established structures?
At present day (2001) as well, a lively academic and policy-oriented discussion flourishes on the influence of (information and communication) technology on society in general, and on its spatial configuration specifically. In this discussion, the effects of ict on regional and urban development and the potential gains of government policy depend on contradictory visions of liberalisation, agglomeration and inertia. Applications of ict infrastructures on first sight enable entrepreneurs to become less place bound: a liberalising effect, summarised as the ‘death of distance’ notion. In this vision, physical distance (to customers, suppliers, labour market reservoirs and services) becomes a less crucial factor for the economic functioning of enterprises (Cairncross 1997, Horan et al. 1996, Borja and Castells 1997). Agglomeration of economic activities is not necessary any more in this scenario; it does not lead to productivity fostering, substantial externalities (Graham 1998). The lay-out of urban regions in the Western world in the past proved to grow both in scale and economic composition by the introduction of new technologies (transport, telephone, automating machinery) in the past (Garreau 1991, Boarnet 1994, Batty and Langley 1994, Kaplan 1998). One would expect that another new wave of technological development, ict, will cause the same spatial effects.
This liberalising view stands in contrast to theories on inertia- and agglomeration induced spatial growth patterns. The presence of cities and urban agglomerations, with accompanying production structures, diversified labour supply and physical and social infrastructures creates externalities that determine where ict-activities will be located (Glaeser et al. 1992, Van Oort et al. 2001). Sometimes this location process depends on existing infrastructures quite literally: trenches for new fibreglass cables are dug parallel to existing electricity cables and conduit-pipes. The new infrastructure enforces the existing one, and further agglomeration of economic activity is induced.
The tension between liberalising and inertia views creates important policy questions. This paper focuses on these from a industrial-organisation economic and spatial perspective. When physical distance as such loses weight in the economic functioning of actors, how can spatial policy gain control over the spatial-economic growth and welfare distribution processes? Are there still economic reasons to cluster activities (in urban regions) in the new digital age? And, when economic growth profits from this, how can spatial policy enhance these processes? For these policy implications, the first main question is how and to what extent ict induces agglomeration or spatial dispersal of economic activities. This is dependent on the embedding of ict in the industrial organisation, within and between firms. The fist part of this paper focuses on this question. In the second part, a translation to possible spatial policy is central. Will it be an uncontrollable odyssey with unanticipated effects of the new digital technology or will there be structura(b)l(e) spatial elements for policy?
ICT, agglomeration and the knowledge economy
Because activities in organisations have to integrated in a firm’s policy and internal organisation, co-ordination of these tasks and functions is at the heart of the firm’s economic process. In general, co-ordination of tasks and functions induces costs. ICT is often hypothesised to make this co-ordination more efficient and less costly. Especially the potential reduction of time-, distance- and transaction costs and efforts leads to more effective and efficient management in this widely accepted view. Time and physical distance become less stringent constraints for economic functioning and production chains of organisation potentially are reduced, either by internal vertical integration and/or external oriented vertical disintegration (Amirahmadi and Wallace 1995, Peck 1996). In fact, this notion resembles arguable similar hypotheses stemming from mechanising and automating (codified information) in previous decades.
The picture becomes more complex when the efficiency of tasks and functions that depend on non-codified or tacit knowledge are related to ict. In table 1 (based on Stumpel and Van der Linde 1998) this especially relates to the elements of quality and innovation, where un-codified, tacit knowledge is very important. The role of ict in this is initially purely supportive. A common vision on this though is that this role is changing: from substituting to facilitating and from effective to innovative. In table 1 the accent changes from top-left to bottom-right. This does not mean that cost-efficiency is not important anymore, but it does mean that the equilibrium balance of relevant aspects clearly changes. More emphasis on networks, facilitated by ict, coincides with a growing importance of knowledge attached to human capital characteristics (Van der Laan 2000).
The shift in attention from substitution towards facilitating is not an entirely new phenomenon, but is strongly related to the societal development of the knowledge economy. Not ict itself, but its role in the knowledge and service economy is crucial for understanding. Knowledge networks within and between organisations are very important in this. The centre of attention in economic theory in this regard is shifting: a transition of models focussing on ‘economies of scale’, via ‘economies of scope’ towards ‘economies of expertise’ (Phelps 1992, Boden and Miles 2000). The knowledge-based organisation differs substantially from the classical organisation. Knowledge is at the core of the enterprise. Labour therefor changes from a cost account into an essential (human resource) investment property. Production processes essentially aim at the creation of immaterial knowledge-structures. Consumer and business relations become part of more personalised networks in which interaction and face-to-face contacts prevail. These (hidden) immaterial assets more and more, and complementary to visible material assets, determine the value of an organisation,
Table 1 The impact of ict on firm performance
We thus argue that the role ict within organisations changes. Central question now is whether this has consequences for the relation between ict and spatial agglomeration theory (Van Oort 2000, 2001, Lambooy et al. 2001). We will shortly discuss four recent, explaining and cross-fertilising agglomeration perspectives, in which ict plays a strong facilitating role. The first perspective starts from the flexible specialisation thesis (Piore and Sabel 1984, Scott 1992, Peck 1996). Since the early 1980’ies many companies have concentrated upon their core competencies. None-core activities are being subcontracted to specialised, flexible firms. When those firms are located in proximity of the larger subcontractor, industrial districts with localised supply-chains emerge. This is accommodated by both ict developments that stimulate the emergence of specialised enterprises taking over the non-core tasks fast and in a flexible manner. Characteristics of the supply-firms are besides their relative small size their supposed innovativeness, efficiency and personal touch on input and output details. In the slipstream of this view, urban development is arguable enhanced by it. New entrepreneurs in facilitating industries often start their business in urban environments, close to customers and consumers; cities then function as incubation milieus (Gerking et al. 2001).
A second, related perspective in ‘modern’ agglomeration theory concerns the urban knowledge economy. Participants and actors in this knowledge economy are usually bound to central urban locations, in which especially face-to-face contacts and ict induced networks are important. Because ict reduces distance-related burdens for many resources, enterprise concentrates more and more on the locational preferences of then most important (and least mobile) production factor: labour. Important in this is the position of the so-called knowledge workers (Florida 2000, 2001, Lambooy et al. 2001, Louter et al. 2001). These are employees that are social competent, network sensitive and co-operation intended. To reduce transaction costs, face-to-face contacts of knowledge workers take place in agglomerated (urban) environments. An important facilitating factor for attracting knowledge workers is, besides the industrial and institutional composition, a diversity of residential and recreational amenities, especially concentrating on an active and cultural climate in environmentally green locations. A third, complementary set of perspectives concentrates on knowledge spillovers in urban environments. In case of knowledge oriented spillovers it is necessary to profit from each other’s proximity in a diversified economic and social structure in which local degrees of competition are very important. This enhanced economic (firm) growth most; growth that would not be present if different types of firms were not concentrated. ICT again facilitates important and crucial ways of communication between and within firms. Empirical research on knowledge spillovers and innovation diffusion though reveals that the (physical) spatial reach of influence is potentially small and that urban borders are only seldomly crossed. Knowledge is apparently most fruitfully exchanged around the corner of the street (Van Oort et al. 2001).
The fourth set of perspectives that explains agglomeration economies is embedded in a geographical, social-institutional approach (most significant put down in Storper 1997). Regions and urban agglomerations in this are assumed (and proven) to be part of a package of social, historical and political factors. Conventions, institutions, learning regions, habits, ethos, the human moment at work: these elements become ever more important for the local functioning of economic activity. But not in a generic way: these elements vary across space in general and across urban agglomeration specifically. In principle these elements are less easy and straightforward for policy to concentrate on than more physical (infrastructural) projects concerning economic activity and its production structure. The spatial reach of Dutch firms seems to be enlarged over time and negative agglomeration externalities (accessibility, rent prices of land) especially seem to affect inner cities of urban agglomerations (Van Oort 2000). Still, economic activity does not spread in all industries at all spatial scales in the same degree, and for some industries and occupation characteristics (the growing industries of business- and personal services) spatial agglomeration is predominantly the spatial trend (Van der Laan and Van Oort, 2000).
The role of ict is not subsidiary to these general spatial-economic trends. Indeed, its forms the glue in facilitating more effectively and efficiently networks between cities and creating agglomeration advantages within cities (Duranton 1999, Gasper and Glaeser 1998, Glaeser 1998). All mentioned modern expressions of agglomeration theory have one thing in common though, and that is the changed meaning of ‘distance’ and ‘accessibility’ compared to traditional spatial-economic paradigms. Proximity to knowledge, as well as social and institutional factors, combined with a diversified industrial and occupational structure is recognised as being of the utmost importance for creating spillovers, growth and interaction.
We therefor argue that the general ‘death of distance’ hypothesis concerning the influence of ict, described at the beginning of the chapter, needs to be reformed into a ‘reform of distance’ hypothesis: from physical to functional distances and from physical accessibility to proximity. ICT potentially enlarges the spatial scale of networks, but it is the fundamental economic embeddedness in agglomeration theory that is essential for understanding the existence and persistence of locally bound networks in the first place. The ‘reform of distance’ hypothesis implies the formation and persistence of agglomerated economic activities complementary to cost advantages; stressing the interchange of codified and implicit knowledge connected to interaction in localised networks. Learning effects are connected to both co-operation and competition within those networks (Scott 1996, Glaeser 1999, Van Oort et al. 2001). ICT does not function as a (deterministic) structuring element on its own account, but is related to and embedded in functions and activities of people, organisations, firms and divisions within firms (Steijn 2001). This makes that, although in terms of effective use of resources, spatial dispersal of economic activities seems to be possible by ict, at the same time economic, institutional and personal local network embeddedness, in which ict also performs a crucial role, restrict this dispersal considerably. In order to stimulate economic growth, spatial policy should have a better eye on especially the extent of these locally bound networks, competitive advantages and circumstances, both in existing as in potential new agglomerations. The latter can be made possible by large firms in growth markets or a cluster of these arge firms that own the power to create their own production environments, as we witnessed in ‘new’ industrial districts like Silicon Valley in recent years.
ICT and the principles of the knowledge economy lead us to adjust ‘the’ agglomeration theory or ‘the’ footloose hypothesis, to put them in a more differentiated perspective. Logically, spatial policy should be reformed similarly, but how? According to the previous section, the use of ict can only be properly understood when focussing on the knowledge economy. This implies that policy should focus more than it does now on organisations and institutions that make up the knowledge economy (Eliasson 2000). Within this economy there is a rising importance of regional and local embedding. This is due to a growing complexity and degree of turbulence, but also because of the growing importance of trust, implicit knowledge and conventions in the economic process (Storper 1997). The growing complexity induces that organisations are only able to gather all relevant information on only a limited number of issues. Therefor enterprises concentrate upon their core business, for which they collect, store and maintain the most relevant information. The rise of turbulence, the fast and often changing market- and competition circumstances, as well technological developments all contribute to a lowering ‘degree of instant, useful information’. This means that organisations can not build on past experiences in all circumstances any more. This increased speed of data and (subsequent) information handling causes that organisations turn for specific knowledge acquirement to their network-dense environment, and especially their local, nearby environment.
The central question now is how organisations, institutions and policy in a spatial-determined framework can establish the right structure that facilitates knowledge spillovers most effectively. And, what role should ict play in this? For this, it is useful to make the distinction between passive and active perspectives of interaction between organisations and regions or locations (figure 1). Traditionally, this interaction is implemented using a passive perspective. Firms and entrepreneurs have localisation demands and regions have locational assets. It is left to the market whether demands and supply actually meet (Hayter 1997). The sole active attitude of regions concern passive stimulus: the supply of physical infrastructure and subsidising investments- or labour costs. This potentially attracts new businesses that enhance (further) economic growth. This passive demand-supply vision inhabits a potential danger though. Many local governments aim at locating as many ict-companies as possible, or think highest priority should be given to opening an inevitable ict- or science parks. Every region wants to inhabit a Silicon Valley. In the predominant industrial society, not that long ago, similar competing considerations led to the construction (‘supply’) of railway, harbour and motorway infrastructures, assuming that it would automatically attract ‘demand’. Many examples, all over the western world, have proven that it does not work that way. But some lessons are hard to learn, and many policy-makers assume that creating infrastructure supply for technology generates demand automatically (Gibbs en Tanner 1997). It bypasses the true demands concerning ict of organisations and economic agents. As was explained above, that is a much more complex question (and answer). Sometimes it is quite difficult to pinpoint what exactly determines demand. ICT- or science parks can be opened, knowledge networks not. Also, investments in ict infrastructure are necessary, but only if regional and local demands justify the amounts of money involved.
In practice, both enterprises and governments at various spatial scales are active participants in this ‘market’. In terms of figure 1, a shift occurs towards the bottom-right quadrant. This implies a larger emphasis on goals, composition and co-operation of institutions (firms, stake-holding groups and governments) for a common development within a region. A thorough knowledge concerning the relevant environments of firms is crucial for that.
Figure 1 Active and passive interaction between the organisation and the region
Building networks on a formal and informal basis between firms, persons and organisations within a certain location or region means that one actively participates in creating the production environment of firms. The spatially embedded network of relations should fundamentally be seen as competitive assets by actors. In organisational theory this is headed under learning organisational development, in which national, regional and local governments co-operate with (semi-) private organisations and stakeholders. In much recent economic-geographical research concerning regional networks and spatial embeddedness these notions of learning are crucial too (Lambooy 2000).
Working in networks, creating regional diversity on almost every relevant aspect, induces that spatial policy in general, and concerning ict specifically, can not suffice using a uniform approach for every individual region. Different functions emerging on different types of locations should be respected in order to enhance and foster economic growth of firms. Policy should me made to measure. Policy choices are dependent upon wishes and possibilities of public and private organisations present in the region. Top-down imposed regional blueprints should be avoided. This insight, that especially concerns ict investments and stimulation, coincides with more general trends and demands concerning spatial policy in western European countries (see Lovering 1999 and Hassink and Lagendijk 2001 concerning experiences in England and Germany). Will a strong regional embedding reduce the position of central governments to ‘wait and see’? No, but the general character of policy does change. Not the physical perquisites (industrial sites, transport infrastructure, parking possibilities or ict infrastructure) or the welfare outcome od regional development (spatial inequality) are central, but the inspiring and facilitating role of institutions and actors that as a common goal try to foster local-economic growth. ICT is in this an essential element.
Summary and conclusions
The change in accents in agglomeration-theory and –policy discussed in this paper is a reflection of a balance between contradictory movements. Liberalisation gives organisations more space of movement and fosters local policy initiatives. Inertia on the other hand determines that developments are usually strictly tight to existing structures, both physical-economic and policy oriented (pipeline policies). The first part of this paper stressed the structural influences of agglomeration economies and externalities on firm performance (productivity growth) due to expertise clustering, knowledge spillovers and network structures. The role of ict in these processes is a facilitating one, enabling a more efficient and effective use of especially labour related resources. The prevailing urban focus of the most important growth industries and occupations within countries of the western world (Fujita et al. 2000) can be explained by this, and it puts the presumed liberalising capacity of information and communication technologies concerning economic activities in its right perspective. Instead of a ‘death of distance’ it is argued that it is better to speak of a ‘reform of distance’, in which there is still (and even predominantly) room for agglomeration externalities.
The second part of the paper, following the first part, logically concludes that initiatives concerning information and communication technologies, albeit not spatially restructuring itself, do need to be fully integrated into active and local oriented policy. Whether central or regional and local governmental agencies are capable of making the right connections and decisions in this is difficult to foresee at this moment. Our general conclusion is that agglomeration externalities, although altered in composition, will still determine the location of economic activities to a large extent in the near future. Economic activity can, dependent on functions and ict-sensitivity, disperse over space. But to what spatial scale remains unclear now. A complete spatial odyssey, experiences of massive footloose firms and industries or extreme shrinkage of space will to our belief, based on an extensive survey of the spatial-economic literature, not be a very realistic scenario for spatial policy to take into (too) serious account.
H. Amirahmadi and C. Wallace (1995), Information technology, the organization of production and regional development. Environment and Planning A 27, pp.1745-1775.
M. Battey & P. Langley (1994), Fractal cities: a geometry of form and function, New York: Academic Press.
M.G. Boarnet (1994), An empirical model of intrametropolitan population and employment growth. Papers in Regional Science 73, pp. 135-53.
M. Boden and I. Miles (2000), Services and the knowledge-based economy. London: Continuum.
R. Borja & M Castells (1997), Local and global. The management of cities in the information age. London: Habitat.
F. Cairncross (1997), The death of distance. How the communications revolution will change our lives. Boston: Harvard Business School Press.
G. Duranton (1999), Distance, land and proximity: economic analyses and the evolution of cities. Environment and Planning A 31(12), pp.2169-2188.
G. Eliasson (2000), Industrial policy, competence blocks and the role of science in economic development. Journal of Evolutionary Economics 10, pp.217-241.
R. Florida (2000), Competing in the age of talent. Environment, amenities and the New Economy. Pittsburg: R.K. Mellon Foundation.
R. Florida (2001), Technology and tolerance: the importance of diversity to high-technology growth. The Brookings Institution Survey Series, June 2001.
M. Fujita, P. Krugman and A. Venables (2000), The spatial economy. Cities, regions and international trade. Cambridge Mass.: The MIT Press.
J. Garreau (1991), Edge cities. Life on the new frontier. New York: Anchor Books.
J. Gasper and E.L. Glaeser (1998), Information technology and the future of cities. Journal of Urban Economics 43, pp.136-156.
S.D. Gerking, D.P. van Soest and F.G. van Oort (2001), Knowledge transfer and the location of new establishments in the Dutch province of Zuid-Holland. Mimeo, University of Wyoming.
D. Gibbs and K. Tanner (1997), Communications technolgy, local economies and regulation theory. Tijdschrift voor Economische en Sociale Geografie, pp.29-40.
E.L. Glaeser (1998), Are cities dying? Journal of Economic Perspectives 12(2), pp.139-160.
E.L. Glaeser (1999), Learning in cities. Journal of Urban Economics 46, pp.254-277.
E.L. Glaeser., H.D. Kallal, J.A. Scheinkman and A. Schleifer (1992), Growth in cities. Journal of Political Economy 100, pp. 1126-52;
S. Graham (1998), The end of geography or the explosion of place? Conceptualising space, place and information technology. Progress in Haman Geography 22, pp. 165-85.
R. Hassink en A. Lagendijk (2001), The dilemmas of interregional institutional learning. Environment and Planning C, pp.65-84.
R. Hayter (1997), The dynamics of industrial location. The factory, the firm and the production system. Chisester: Wiley.
T.A. Horan, B. Chinitz & P. Hackler (1996), Stalking the invisible revolution. The impact of information technology on human settlement. Claremont: University Research Institute.
R..D. Kaplan (1998), An empire wilderness. Travels into America’s future, New York: Random House.
L. van der Laan (2000), Making room for ict. Economisch Statistische Berichten, p.D27-31 (in Dutch).
L. van der Laan en F. G. van Oort (2000), The urban labour market paradox revisited. Rooilijn 3, pp.116-122 (in Dutch).
J. Lambooy (2000), Learning and agglomeration economies: adapting to differentiating economic structures. In: F. Boekema et al., Knowledge, innovation and economic growth. Edward Elgar, Cheltenham, pp.17-37.
J. Lambooy, E. Nagengast, N. Raat en L. Veldkamp (2001), Spatial effects of ict in the Netherlands. RPD-report, ’s-Gravenhage (in Dutch).
J. Lovering (1999), Theory led by policy: the inadequacies of the ‘new regionalism’, International Journal of Urban and Regional Research, pp.377-395
P.J. Louter et al. (2001), Making room for the digital economy. Inro-TNO, Delft (in Dutch).
F.G. van Oort (2000), Unity in diversity? ICT and urban development. Stedebouw en Ruimtelijke Ordening, pp. 19-25 (in Dutch).
F.G. van Oort, D.P. van Soest and S. Gerking (2001), Dynamic information exernalities and employment growth in the Netherlands. Mimeo, Katholieke Universiteit Brabant, Tilburg.
J. Peck (1996), Work-place. The social regulation of labor markets. New York: The Guilford Press.
N.A. Phelps (1992), External economies, agglomeration and flexible accumulation. Transactions of the Institute of British Geographers 17, pp. 35-46.
M.J. Piore and C.F. Sabel (1984), The second industrial divide. Possibilities for prosperity. New York: Basic Books.
A.J. Scott (1992), The collective order of flexible production agglomeration: lessons for local economic development policy and strategic choice. Economic Geography 68, pp.219-233.
A.J. Scott (1996), The craft, fashion and cultural-products industries of Los Angeles: competitive dynamics and policy dilemmas. Annals of the Association of American Geographers, pp.306-323.
B. Steijn (2001), Working in the information society. Assen: Van Gorcum (in Dutch).
M. Storper (1997), The regional world. Territorial development in a global economy. New York: The Guildford Press.
J.W. Stumpel and R. van der Linde (1998), ICT and the economy. Ministry of Economic Affairs, ’s-Gravenhage (in Dutch).