New Directions in Economic Geography
Bernard Fingleton (Editor)
Edward Elgar, Cheltenham (2007), 384pp, £75.00 (hbk), ISBN 978-184542373-5.
Reviewed by Frank van Oort,Urban Economics and Spatial Planning, Utrecht University, The Netherlands.
This volume discusses recent new directions in geographical economic research. It tries to bridge the disciplines of economics and geography. It offers good overviews of core concepts in a series of sub disciplines, ranging from the New Economic Geography (NEG), urban economics, transport economics, complexity theory and international economics and FDI. It is critical on theoretical and empirical achievements in geographical economics so far. It presents future new directions for geographical economic research. Three general impressions of the book remain after reading: first, that this is a good and genuine attempt to cover all these sub disciplines aspects in a common spatial-economic framework (namely that of geographical economics), second, that theorizing and empirical research are still miles apart and that much work remains to be done on all possible aspects, and third, that this is worthwhile doing. A very interesting volume indeed, recommended reading for everyone interested in theorizing space in economics or working in the empirical spatial-economic research arena.
The book takes the New Economic Geography (better labeled geographical economics) as a starting and reference point. This body of theory is first and foremost an outcome of the creative imagination of economists and not geographers, and is rooted in international trade theory. Summarizing its history, it is clear that prior to the development of new trade theory, traditional international trade theory was largely unable to explain either intra-industry, intra-national or intra-regional regional trade. At the same time, gravity models suggested that most trade tended to be localized. The development of new trade theory based on the Dixit-Stiglitz (1977) modeling framework, subsequently lead to renewed interest in both localized and intra-industry trade. These developments in international trade theory in turn lead to a renewed modeling interest in spatial economics in the form of New Economic Geography, and regional economics as a whole subsequently experienced a resurgence via a combination of the developments in both New Economic Geography and also New Growth theories. It is important to notice that both New Growth theory and New Trade theory predate New Economic Geography. In both of these strands of literature the dominant analytical approach is the modeling of imperfect competition and increasing returns to scale within the monopolistic competition framework of Dixit and Stiglitz, in which utility is a function of variety. New trade theories now allowed for the modeling of trade flows within a general equilibrium framework in which the structure of demand and supply is endogenously determined. Krugman (1991) first applied this modeling framework to the question of geography under conditions of economies of scale and labor mobility, and reinterpreted Marshall’s famous principles of externalities as stemming from the benefits of the pooling of the local labor supply and the demand for specialized non-tradable inputs. In these models, spatial concentration and dispersion were seen to emerge as a natural consequence of market interactions involving economies of scale at the level of the individual firm, with many of the results generated by these models being reminiscent of the results of central place theory and the rank-size rule (Fujita et al. 1999). The cumulative causation characteristics of these models is in many ways akin to the processes described amongst others by Pred (1977) and in this respect the Krugman-Fujita-Venables work builds on most of the standard location theory (McCann and Van Oort 2008). These types of arguments therefore provide some additional possible explanations for systematic variations in competitive advantage across regions and why it is that certain regions are able to maintain and even reinforce their advantages over other regions, once certain locations have taken a lead in a particular activity.
Neary (2001) and Martin (1999) formulated serious criticisms, stressing that theory and empirics of the New Economic Geography are both underdeveloped, let alone that they are connected. The chapters in the book all witness this; the contributions all highlight important pitfalls, and can be seen as a constructive critique on the New Economic Geography. The editor, Bernard Fingleton, in an earlier paper in Environment and Planning A in 2000 already formulated that this work is to bridge economics with geography and theory with empirics – and thus to find a ‘third way’ of reasoning in-between the two disciplines. Where this ‘third way’ parts company with what remains of traditional New Economic Geography is the emphasis it gives to realism – in other words, important real-life outcomes, structures and processes are “not assumed away because they get in the way of formal modeling, but are incorporated because at the end of the day theory is about explaining reality, and is somewhat sterile and fairly useless when treated as an end in itself” (p.2-3).
The chapters are written by economists with an open mind for economic geography. Some chapters bring the empirical limitations of NEG-models explicitly to the fore. A simple simulation model based on New Economic Geography – although partial and incomplete - is able to replicate reality in wage levels in the UK, but the conditions for agglomeration appear to be crucial in Bernard Fingleton’s analysis in the first chapter. He also tests NEG-theory against the competing theory derived from urban economics in the third chapter. The principal finding is that NEG is no better than urban economic theory as a predictor of wage variations across European regions. Gianmarco Ottaviano compares two NEG-models – one based on factor mobility and one on vertical linkages – and he concludes that they lead to the same key predictions. This then leads to the recommendation that the two theories somehow should be joined. In the chapter by Steven Brakman and Harry Garretsen, a test is carried out whether NEG-models are able to predict break points in economic integration for policy purposes. Their confrontations of NEG empirics with NEG theory points to serious limitations of current NEG research. And although, as mentioned above as well as in the chapter by Mark Roberts and others on productivity growth, the New Economic Geography in itself is no growth theory (but rather a static-comparison model), Fabio Cerina and Francesco Pigliaru clearly show that development trajectories in NEG-models crucially depend on very restrictive values of some parameters in their model. The chapter by Bernard Fingleton and Phil McCann questions the assumption of iceberg transport costs in NEG modeling. They convincingly argue that these are unrealistic, in the sense that they do not allow for any economies of distance and of the scale in the transportation of goods and information, as is shown to be the case in the growing empirical literatures on the geography of innovation (Acs 2002) and on spatial econometrics, in which the definition on distance and network weight matrices is the most crucial element in modeling (Anselin, Florax and Say 2004). The chapter by John Dewhurst and Phil McCann touches upon the modifiable areal unit problem (MAUP), a much neglected aspect in spatial economic modeling. From the early 1990s onwards, urban economists and NEG-modellers have shown an increasing interest in agglomeration externalities that endogenously induce localized economic growth. This development can mainly be ascribed to the failure of orthodox economics to give appropriate explanations for the variation in the wealth and poverty of cities and regions. Many geographers and economists claim to be part of this ‘new geographical turn in economics’ (Martin 1999), but one of the main criticisms on the empirical literature on agglomeration is that it pays little attention to the spatial configuration of cities and regions and the geographical scale of agglomeration benefits. Although every research on the benefits of agglomeration starts off with the notion that ‘space matters’, it is strangely enough the same ‘space’ that is inadequately dealt with in most formal mathematical models. More explicit indictments take in that most urban economic studies on agglomeration ‘are space neutral’ (Van Oort 2004: 7), ‘implicitly model cities as a club’ (Rosenthal and Strange 2003: 377), and, above all, insinuate that ‘similar rules apply at all spatial scales’ (Overman 2004: 513). We can distinguish between three major analytical issues to which the present empirical literature on agglomeration pays insufficiently attention: spatial dependence, spatial heterogeneity and the MAUP. Although a growing body of literature proves that spatial dependence and heterogeneity can be dealt with by means of spatial econometric techniques (Anselin et al 2004), little is known about the scale sensitivity of research results for variations in the initial unit of spatial analysis. Despite repeated warnings of the potential influence of spatial composition and aggregation effects in the related geographical literature (e.g. Wrigley 1995), the spatial economic modeling tradition on agglomeration externalities has not taken up this issue seriously. Dewhurst and McCann show that empirical research (in their case on the relation between specialization and urban hierarchy, in which smaller cities in the UK tend to be more specialized) indeed is dependent on the initial scale of analysis. It must be said that this does not comprise a full treatment of the MAUP, since that consists besides scale effects also of zoning and gerrymandering effects, and hence more research on this is needed. A solution to these problems can be the focus on micro-level firm data (Arbia 2001), a suggestion in different contexts also made by Anna Soci in her chapter on FDI-investments in regions and by Paul Plummer and Eric Sheppard in their plea for complex spatio-temporal dynamics of firms. The latter chapter also gives a intriguing conceptual account of complexity theory in a spatial context. Finally, the contribution by Alessio D’Ignzaio and Emanuele Giovannetti focus on the spatial effects of ICT, and conclude that proximities (either physical or network related) still play a role in the behavior of internet providers, and Andres Rodriguez-Posse and Ugo Fratesi show that the returns of European structural policies contribute – unintended - to regional economic divergence and greater economic agglomeration.
The recommendations for further research in the chapters merge with the two main questions that arise from urban economic theory (Cheshire and Duranton 2004, Henderson and Thisse 2004). The first asks what the micro-foundations of agglomeration economies are, and whether insight in this question will provide useful building blocks for further modeling and theorizing. The second asks who benefits from agglomeration economies, and to what extent. Rosenthal and Strange (2004) review a large number of studies that use cross-section data on cities (typically but not exclusively in the US) to relate productivity to city size. Findings vary according to the extent to which researchers are able to control for the quality of inputs – capital stock and the skill of the labor force – and typically yield the result that doubling city size increases productivity by an amount in the range 3- 8%, i.e. an elasticity of productivity with respect to city size of between 0.04 and 0.11. Is this the true figure? Who benefits from it? Which industries, types of firms, in which regions? This is essential information for our understanding of urban economic and NEG-theories. The chapters in the book themselves do not provide the answers. As the back of the dust jacket rightly states, the volume can to some extent be regarded as part of the inevitable creative destruction of NEG theory. But as long as we are looking for answers, the theoretical and empirical criticism in the volume is constructive and needed more.
References
Acs, Z.J. (2002), Innovation and the growth of cities, Cheltenham: Edward Elgar.
Anselin, L., R.J.G.M. Florax and S.J. rey (2004), Advances in spatial econometrics. Heidelberg: Springer Verlag.
Arbia, G. (2001), Modelling the geography of economic activities on a continuous space, Papers in Regional Science, 80: 411-424.
Cheshire, P. and G. Duranton (2004), Recent developments in urban and regional economics. Cheltenham: Edward Elgar.
Dixit, A.K. and J.E. Stiglitz (1977), ‘Monopolistic competition and optimum product diversity’. The American Economic Review, 67, pp. 297-308.
Fingleton, B. (2000), ‘Spatial econometrics, economic geography, dynamics and equilibrium: a third way?’. Environment & Planning A 32, pp. 1481-98.
Krugman, P.R. (1991), ‘Increasing returns and economic geography’. Journal of Political Economy 99, pp. 483-499.
Martin, R. (1999), ‘The new “geographical turn” in economics: some critical reflections. Cambridge Journal of Economics 23, pp. 65-91.
McCann, P. & F.G. van Oort (2008), ‘Theories of agglomeration and regional economic growth. A historical perspective’. In: R. Capello & P.
Nijkamp (eds.), Regional dynamics and growth: advances in regional economics. Cheltenham: Edward Elgar (forthcoming).
Neary, J.P. (2001), ‘Of hype and hyperbolas: introducing the new economic geography’. Journal of Economic Literature 39, pp. 536-561.
Overman, H.G. (2004), Can we learn anything from economic geography proper?, Journal of Economic Geography, 4: 501-516.
Pred, A. (1977), City systems in advanced economies. Past growth, present processes and future development options. London: Hutchinson.
Rosenthal, S.S., Strange, W.C. (2003), ‘Geography, industrial organization, and agglomeration’, The Review of Economics and Statistics, 85: 377-393.
Rosenthal, S.S. and Strange, W.C. (2004), ‘Evidence on the nature and sources of agglomeration economies’. In Henderson, J.V., Thisse, J.F. (eds), Handbook of Regional and Urban Economics: Cities and Geography, Volume 4. Amsterdam: North Holland, pp. 2119-2179.
Van Oort, F.G. (2004), Urban growth and innovation. Aldershot: Ashgate.
Wrigley, N. (1995), ‘Revisiting the Modifiable Areal Unit Problem and the Ecological Fallacy’, in Cliff, A.D., Gould, P.R., Hoare, A.G., Thrift, N.J. (eds), Diffusing Geography. Oxford: Blackwell, pp. 49-71.
Frank van Oort is professor in Urban Economics and Spatial Planning at Utrecht University, The Netherlands. He also works for the Netherlands Institute for Spatial Research (RPB) in The Hague.
(Added 10 January 2008)