Defining Appropriate Economic Space

Spatial analysts have shied away from defining optimal economic space. Roger Vickerman (1980: 165) states that “the search for optimality in models of spatial economies is probably as fruitless as that for optimality in the economies themselves” because the models are too restrictive or too simple. As “optimal spatial planning is not a practical proposition in operational terms”, he suggests that rather than “start with the rigid framework of equilibrium” the task is ”to build up our picture of the spatial economy by a detailed consideration of the interactions and linkages between decisions and decision-makers.”

Almost three decades later, Vickerman (2007, pers. comm. 2008) admits that there is a better understanding of the impact of infrastructure on regional economic cooperation, but his basic message is still valid: do not assume that investment in infrastructure will solve all problems. What Vickerman omits to say, however, is that it does not make sense to try and identify ‘optimality' in the conditions of rapid growth and structural change that pertain across Asia. All that can be done is to determine whether a particular economic space is appropriate for the purpose at hand.

National governments use cost/benefit analysis to rank spending priorities, including where to locate investment, what kind of investment, and how much investment. The problem is that the usual national calculus omits cross-border effects or externalities (spillovers) and underestimates the potential of trade and foreign direct investment.¹ This gives rise to a problem of negative externalities or spillovers.

In large countries such as the People's Republic of China (PRC), India and Indonesia, international spillovers are apparent only in their border areas. In smaller countries, however, spillovers have a proportionately larger effect. Apart from the tiny Pacific Island economies, this is also the case, for example, in Singapore, Malaysia, and Cambodia. Thus, the key issue is not optimality but what is appropriate to the case in question—Mainland Southeast Asia, Island Southeast Asia or the Pacific Islands.

These instances are sufficient to highlight that a ‘seamless' Asia is a long way from realization. Asia is not only separated by oceans and deserts, and split by different cultures and linguistic groups, but it is also divided by national boundaries. The boundaries are points at which political jurisdictions change and where international transit trade data are collected. They are associated with barriers to the flow of goods, not only stemming explicitly from tariffs and quotas, but also arising implicitly from inconsistent standards and irritations linked with border crossings. However, these boundaries pose obstacles to trade even where no formal barriers exist (Fujita, Krugman and Venables, 1999). National boundaries also pose obstacles to movements of people that persist well after formal barriers to trade in goods have been lowered.² Even if we dispensed with national borders and considered trade flows across unbounded economic space, distinct regions of industrial specialization would still evolve of their own accord and exhibit a distinct centre and periphery (Fujita pers. comm., 1999).

Urban agglomeration is becoming increasingly important due to the globalization of the world economy. The process is examined as a general equilibrium analysis between agglomeration economies stemming from inter-firm linkages and agglomeration diseconomies arising from immobile factors. Specifically, Masahiko Fujita, Paul Krugman and Anthony Venables (1999) give special attention to the impacts of increasing returns and degree of interactive activity, and the effects of transport and communications costs on urban concentration and dispersion. As outlined in Figure 1 [ PDF 33.5KB | 1 page ], the main findings of their theoretical exercise are that tensions between centrifugal and centripetal forces shape the spatial economy.³

1. The dispersion of firms arises when transport and communication costs are sufficiently low and immobile factors such as land and, in some cases, labor are important centrifugal forces.
2. Structural transition from dispersion to concentration occurs when transport and communication costs rise to intermediate levels and firms engage in backward and forward linkages to create the circular logic of agglomeration (i.e., producers want to locate close to their suppliers and customers).
3. The re-dispersion of firms (i.e. deconcentration and decentralization) occurs when transport and communication costs have declined sufficiently over time to allow easy access to other firms and consumers, and agglomeration is no longer important due to lower factor prices and weaker competition.

Even when transport and communications costs fall, the spatial structure marked by agglomeration may persist over decades due to interdependence between the locational decisions of firms.

The welfare and policy implications of urban concentration and dispersion are addressed by Takatoshi Tabuchi (1998) after synthesizing William Alonso's (1964) work stressing intra-city transport costs and Paul Krugman's (1991) paper emphasizing inter-regional transport costs, Tabuchi sees agglomeration policies as being desirable in the process of urbanization, but unnecessary in the case of re-dispersion, which is attained without government intervention (see also Krugman, 1995). Where interregional transport costs are high, it follows from a social welfare viewpoint that dispersion is worse than agglomeration and that a decentralization policy, often conducted in regional planning, is not justified.

As Asia becomes more integrated and when critical thresholds are reached, falling transport costs may trigger an abrupt change in economic geography, disrupting any temporary equilibriums that may have been attained. As in the recently emergent PRC, shifts in regional specialization would be prompted, leading to the coalescence of new economic regions. Should transport costs fall, changes in the distribution of industry between centre and periphery, or a reversal of their respective positions, may occur.

Economic space is not only unbounded but is also uneven as reflected in the concentration of flows in corridors between main cities; multi-dimensional in the sense that there is an array of air, sea, land, tele- and financial spaces that have to be integrated; and, discontinuous with city cores linked by telecommunications being more adjacent economic spaces than physical hinterlands (see Box 1 [ PDF 102.7KB | 1 page ]). These observations underline the significance of urban hubs and (international) inter-city corridors. It is important not to become too preoccupied with cross-border land crossings and to focus on inter-city connections. Rural areas do not aggregate demand for airlines, container ships and telephone traffic; their place in the articulation of national and international networks involves the collection and distribution of commodities at thin densities of traffic and hence imposing only modest demands on infrastructure.

In Asia the capital or main cities are the largest agglomerations of economic activity and the main generators of cross-border flows, primarily because market forces are irresistible and largely impervious to any government attempts to slow them down. A policy of re-dispersing activities from these cities is very difficult to promote because it runs counter to market forces and takes decades to realize. In short, it is seldom possible to develop a new main city in a place chosen by policymakers. Removal of price or investment distortions will correct imbalances between capital or main cities, small towns and rural areas. However, as illustrated by Ben Higgins and Donald Savoie (1995) in regard to Malaysia, planners have had little success in altering the spatial distribution of economic activity to accord with the stated goals of national policy. Rather than leave the location of spread effects to the market, letting them occur where they would without intervention, planners sought, ineffectually, to direct them to certain areas to benefit particular societies.

These observations do not mean that all available resources should be poured into capital or main cities.⁴ The proposition reflects that on the whole capital or main cities, with the exceptions of Singapore and Kuala Lumpur, suffer from public underinvestment. Private investment has been focused on shopping malls rather than infrastructure (Rimmer and Dick, 2008). Many cities have had no major infrastructural additions since the colonial era, and the city and its networks have been overwhelmed by the consequences of rapid economic growth. The history of the world suggests that urbanization goes in tandem with industrialization.

The capital or main cities in Asia have coalesced with industrialization. Since the 1980s industrialization has shifted the economic centre of gravity away from extensive agricultural hinterlands to the main cities (Dick and Rimmer, 2003). Since then the immediate hinterland and associated economies of agglomeration have transformed the main cities into extended metropolitan regions typified by Greater Bangkok, Greater Jakarta and Greater Manila. Not only are these mega-city regions once again closely linked with the world economy but they are much less dependent on their agricultural hinterlands, becoming the new enclaves. In turn, the main cities are also the gateways that may constitute a high proportion of national economies (e.g. Bangkok in Thailand, Manila in the Philippines and Kuala Lumpur in Malaysia) and are the focus of multimodal corridors.

Download this Paper [ PDF 1.2MB| 38 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.