Regional Science Research Paper

This interest in the region as a subject of study had important and immediate political and intellectual antecedents in the United States. In 1927 the regional plan for New York considered economic factors in metropolitan growth and spatial form, largely through the work of Robert Haig. In 1933 the Tennessee Valley Authority became the nation’s first major regional development program, arguably since the western expansion. In 1935, asserting that policies of national development depend on regional factors, the US National Resources Committee asked ‘modern students of regional science’ (10 regional geographers and two regional sociologists) to identify the best type of region to use for regional planning and development. In 1942, Alvin Hansen, perhaps the leading American Keynesian economist and a Harvard professor, advocated a comprehensive postwar expansionary program of regional development on the premise that the effective utilization of resources in any region makes the whole nation richer.

Two organized, interdisciplinary efforts with very different intellectual roots had focused on regions. In the 1920s Lewis Mumford and his colleagues in the Regional Planning Association of America saw cities as congested and unmanageable. They advocated a mode of thinking and procedural methods aimed at seeing the region as a whole—integrating city and region, and utilizing, not destroying, the natural environment. Closer to the regional scientists because of their emphasis on both science and the region and their scholarly, university base were the southern regionalists. Centered at the University of North Carolina, they were well established by 1931 in terms of research achievements, agendas, and infrastructure. Howard Odum, the acknowledged leader of the effort and a sociologist, argued in 1945 that the nation owes the fullest development of the South not only to the South but also to the nation itself. Similar in ambition but predating the Regional Science Association by a quarter century, Odum advocated a regional approach to the scientific study of human society, arguing that it would entail the integration of method and theory, require the joint effort of the traditional social science disciplines, and create fundamental new understanding of all society, not just its regions.

Members of the Regional Science Association from the outset debated whether regional science would become a new discipline or an interdisciplinary activity. Walter Isard, widely recognized as the father of regional science, was the leading proponent of the disciplinary view. In 1956, he became an economics professor at the University of Pennsylvania, where he formed the Department of Regional Science in 1958. By then he had published 23 papers in the leading economics journals (acknowledging the help of Alvin Hansen in the first five). Yet the Regional Science program would go far beyond economics in its scope and impact. In 1960 the first Ph.D. in Regional Science was awarded to William Alonso, later regional planning professor at Berkeley and Saltonstall Professor of Population Studies at Harvard. Subsequent doctoral recipients took positions in economics, engineering, geography, planning, and other university faculties as well as public agencies and consulting firms.

Isard’s prediction that regional science programs would emerge from geography departments and achieve wide recognition before 2000 was not realized. In the 1970s Cornell University established the second Regional Science Ph.D. program, and the University of Illinois initiated a multidepartmental program requiring completion of the Ph.D. in a participating department. By 1976, regional science degree programs or specializations existed at 14 universities internationally. The lone department devoted exclusively to regional science remained the one at Pennsylvania until it was closed during university retrenchment in 1994.

Among the early challengers of the disciplinary view was Lloyd Rodwin, a planning professor at the Massachusetts Institute of Technology. He envisioned the social sciences developing more systematic spatial studies, geography becoming more analytical, and regional science becoming ‘a generic name for regional studies.’ This vision has become reality, but not fully. Regional science has become a generic name for a certain kind of regional studies, one whose origin is in mathematical abstraction and quantitative modeling.

Today regional science is an international, interdisciplinary endeavor. It brings together economists, geographers, planners, and others whose research has a regional dimension. It is the confluence of many streams, but arguably also the origin of a few. Each discipline brings different theoretical and methodological approaches to regional science, and each typically has different substantive concerns within a given topic. Migration, for example, interests several disciplines. An economist might focus on migration behavior as a way of measuring quality of life in cities, a demographer on regularities in the age pattern of migration, a geographer on turnarounds in core– periphery migration, a planner in forecasting migration for land use and transportation modeling, a sociologist on the effect of kinship links on migration destinations, and a real estate specialist on the relationship between housing prices and migration.

The numerous scholarly regional science journals and annual conferences create effective forums for exchanging ideas among disciplines. Although dominated by the abstraction and modeling inherited from economics, these forums are eclectic and open. They have been helpful incubators for scholarly groups who went on to create their own forums, be they input–output analysts, spatial social theorists, urban economists, or transportation modelers. Collectively regional scientists actively borrow ideas from a wide range of sources and combine, adapt, and extend them to regional research.

The recognition Isard anticipated has materialized in many ways other than through a proliferation of university departments. The Regional Science Association became the Regional Science Association International (RSAI), an umbrella organization over- seeing three major supraregional organizations, the North American, European, and Pacific coordinating councils. Most countries in Asia, Europe, and North America have active regional science associations, and there are several multinational ones, including the British–Irish, French-language, German-speaking, and Australia–New Zealand/organizations. Most sponsor annual meetings, and several have journals.

A widely accepted notion is that almost all regional scientists have dual identities, economist and regional scientist, geographer and regional scientist, and so on. For many, the regional science meetings and journals provide the best audience for honing and distributing their research. Others’ participation ebbs and flows with changes in their immediate research agendas. No complete census of this fluid regional science com- munity exists. The associations, including those in China, India, and Brazil, may have over 10,000 members combined. RSAI itself has about 2,000 dues-paying members annually.

Regional science journals have become numerous and well established. Five international ones with regional science in their titles are now published by major commercial presses: Annals of Regional Science, International Regional Science Review, Journal of Regional Science, Papers in Regional Science, and Regional Science and Urban Economics. Echoing the debate a half century ago about the name ‘regional science,’ some regional science associations leave it out of their journal titles, for example, Journal of Regional Analysis and Policy, Review of Urban and Regional Development Studies, or Review of Regional Studies.

Book series, position announcements, honorary degrees, and external research prizes are other marks of regional science’s role within the social sciences. Noteworthy, too, are the National Science Foundations program in Geography and Regional Science in the US and similar ones in other countries. Perhaps the most spectacular recognition of regional science occurred when the Emperor and Empress of Japan personally welcomed participants to the RSAI World Congress in 1996.

1. Subject Matter

The international regional science community studies a wide range of topics, sometimes ambitiously defined as those social, economic, political, and behavioral phenomena that have a spatial dimension. A common, narrower view groups regional science research under three themes: location theory, methods of regional and spatial analysis, and regional development and policy.

Location theory is the study of what social and economic activities are located where and why. Traditionally, emphasis has been on the location of firms and households. The latter considers inter-regional and international migration and labor markets, as well as intrametropolitan residential mobility and housing markets. Early contributions focused on the role of transportation costs and proximity to suppliers and markets. Recognizing interdependencies of behavior focuses attention on agglomerations or clusters in space, as well as questions of land use, urban form, housing markets, city size, hierarchies among and systems of cities, market thresholds, and market boundaries. Recognizing interactions among activities and individuals distributed over space raises questions about commuting, trade flows, congestion, transportation, other infrastructure, facility location, taxes, user fees, local public finance, and diffusion of innovations and growth.

The phrase, methods of regional analysis, describes a growing set of techniques used to identify, analyze, or forecast key regional characteristics or changes. Among the methods are three invented to study national economies but adapted to regions. The inventors received Nobel Prizes in Economic Sciences: Wassily Leontief (1973) ‘for the development of the input–output method and for its application to important economic problems,’ Lawrence Klein (1980) ‘for the creation of econometric models and their application to the analysis of economic fluctuations and economic policies,’ and Sir Richard Stone (1984) ‘for having made fundamental contributions to the development of systems of national accounts.’ Regional scientists have adapted and extended techniques from demography. The cohort-component model focuses on the components of population change and their probabilities (births, deaths, and migration), grouping the population into cohorts by age, sex, and often race or ethnicity. Regional scientists borrowed the gravity and entropy models directly from physics and found that they described certain aggregate human behavior well. Migration, trade, transportation flows, and other interaction between two places tend to increase with the product of their masses (e.g., populations) and decrease with distance between them. Regional scientists have also adapted and applied optimization techniques to contexts in which spatial interaction is important, such as transportation networks, location of facilities, and land use.

Regional scientists use statistical methods extensively. Yet here, too, important adaptations and extensions permit the study of spatial and regional phenomena. Geographers, in particular, have extended standard statistical concepts to situations where observations have locations, for example, to draw random samples over space or to determine the likelihood that an observed locational pattern could have occurred by chance. Spatial statistics, spatial econometrics, geographical information systems, and other methods of spatial statistical analysis have become major parts of regional science—for economists, planners, engineers, and others as well as geographers. In the case of statistical methods widely used to study travel demand, residential location, migration, and other behavior with a spatial dimension, the pioneer was an economist, Daniel McFadden, who won the Nobel Prize in 2000 ‘for his development of theory and methods for analyzing discrete choice.’

Regional development and policy focuses on why some regions grow and prosper while others lag and even decline. Research ranges from theories of growth and development to evaluation of specific policies. Persistent questions deal with the divergence and convergence of regional growth rates and incomes, the relationships between core regions and peripheral regions, and the geography and timing of the spread of growth. Other long-standing questions deal with regional labor markets, including the relationship between growth and unemployment and the persistence of wage differences among regions. Key policy issues concern people prosperity vs. place prosperity (should the focus be on helping people directly, for example, through education or out-migration incentives or on helping people indirectly by making the region more competitive through infrastructure, tax incentives, or other actions), ways of making lagging regions more competitive and prosperous, the rationales for regional policy, and the roles of different levels of government.

Policy evaluation has attracted much attention as well. Commonly the objective is to measure the effects of various (nonregional) national policies on regions, for example, free trade treaties, defense spending, or tax reductions. Other times the objective is to evaluate regional policies, for example, by estimating what would have happened to lagging regions in the absence of the regional support programs. Sometimes the emphasis is on comparing alternative regional plans or policy instruments before they are implemented, generally using cost-benefit, cost-effectiveness, or goals achievement methods augmented to include policy objectives, distributional issues, and hard to quantify outcomes.

2. Intellectual Progress

Two easily contested, unavoidable decisions must be made before discussing the intellectual progress of regional science. The first is whose contributions should be counted. The open, fluid nature of the regional science community makes this decision difficult, and the early claims to disciplinary status make it controversial. The perspective here is that the idea itself is more important than the primary tribal self-affiliation of its originator. The key criterion is whether the idea is part of the scholarly conversation in regional science meetings and journals. Mention of an individual’s contribution does not imply that the person considers him or herself a regional scientist, although each person mentioned has participated in regional science meetings or published in regional science journals. Debates about whether particular intellectual contributions were initiated by regional scientists, borrowed by regional scientists, or adapted by regional scientists, and who is really a regional scientist are fruitless. The gravity model provides an early case in point. One observer described it as work that may clearly be termed regional science, while another argued that it had been pioneered by outsiders and applied in many disciplines. By far the most interesting fact here is how understanding and use of an idea have evolved.

The second decision concerns the scope of regional science. Something narrower than human activities and phenomena with an important spatial dimension must bound this discussion, but many offer too incomplete a view. Some describe regional science as a toolbox for practical analysis that lacks rigorous theory, while others decry it as a theoretical nevernever land with too little realism. The path chosen here recognizes Isard’s deliberate efforts to nurture the field of regional science. He edited several of its journals, commissioned papers, supported dissertations, organized meetings, and shaped an influential book series. Three of those books frame a useful way of discussing regional science: Isard’s Location and Space-Economy (1956), the Isard-led collaborative effort, Methods of Regional Analysis (1960), and a reader edited by John Friedmann and William Alonso and dedicated to Isard, Regional Planning and Development (1964). The books’ foci—theory, method, and policy—frame the discussion of intellectual progress well.

2.1 Location And The Spatial Economy

The quest here is to model mathematically and show graphically both the determinants and consequences of individual location decisions. Regularities in spatial patterns are readily observable, among them, the emergence and persistence of large cities, clusters of related industries, relatively homogenous residential areas within cities, and regularities in city size distributions. The goal, chiefly of economists who work in this tradition, is to show how these agglomerations materialize from the interaction of self-serving decisions of individuals and firms. Isard’s book drew on the pioneering German contributions by Johann Heinrich von Thunen, Alfred Weber, August Losch, and Walter Christaller.

Two works from the birth period of modern regional science were extraordinarily seminal. Alonso’s dissertation, Location and Land Use, focuses on transportation costs and its role in residential decisions and the resultant land use patterns within a metropolitan area. Concentric rings of land use emerge from the behavior of utility maximizing individuals. Alonso’s starting point like von Thunen’s is the existence of an agglomeration, in this case the central business district of a monocentric city. Charles Tiebout, a founding member of the Regional Science Association, argued that ‘people vote with their feet’ by choosing locations within the metropolitan area that offer the combination of revenue and expenditure patterns that appeal most to them. Thus, spatial mobility among diverse communities within an urban area provides a solution to the collective consumption problem posed by Samuelson and another mechanism by which individual preferences create spatial variation. Much of today’s field of urban economics builds on these two contributions.

A contemporary contribution of this genre is The Spatial Economy (1999) by Masahisa Fujita, Paul Krugman, and Anthony Venables, which they describe as a continuation and possible validation of Isard’s project. It derives agglomerations instead of positing them as a starting point (e.g., Alonso’s central business district around which commuters organize themselves). In the ‘new economic geography,’ as Krugman calls it, agglomerations result from the tension between centripetal forces that promote concentration of economic activities and centrifugal forces that oppose it. Of key interest is when a spatial concentration of economic activity is sustainable, and when a system without spatial concentration breaks down and concentration emerges. Combining formal theoretical models that include increasing returns to scale and computer simulations to study nonlinear dynamic properties, this work links urban, regional, and trade economics in a way that exceeds the dreams of the formal modelers of a half-century ago. As was the case then, however, many geographers and others within the regional science community do not embrace this formal mathematical, theoretical mainstream economics approach. They are far more interested in the social, cultural, and institutional factors involved in spatial development and in the study of real places and real regional histories in contrast to abstract ones. The new Journal of Economic Geography hopes to encourage dialogue among the groups.

2.2 Methods Of Regional Analysis

A very different early quest of modern regional science was the creation of practical tools for the study of real places, and here regional science has had spectacular success. The input–output model attracted early attention. A series of regional input–output articles appeared in the Review of Economics and Statistics between 1951 and 1971. Their foci ranged from the theoretical specifications of the models and the definitions of concepts to studies of the nature of metropolitan economies, the location of a major steel complex, and the effects of arms reduction. This initial period culminated in the creation of two landmark regional input–output models constructed from survey-based data, the Philadelphia model of Isard and Thomas Langford and the West Virginia model of William Miernyk. The next three decades focused on building and using models, with attention paid to the treatment of households and consumption, economic–demographic linkages, and strategies for building the models in a cost-effective way. Here too progress exceeded the dreams of the 1950s, with regional input–output models now widely used for impact studies in the private and public sectors in many countries of the world. In fact, this method of regional analysis has become an industry with data and software readily available from commercial firms, most notably the IMPLAN system within the United States.

Econometric modeling similarly saw astonishing progress from the efforts of pioneering scholars to the commonplace of professional practice. Lawrence Klein sketched out how regional econometric models might be developed as satellites of national models, and his student Norman Glickman developed a fully operational landmark model of the Philadelphia economy. For roughly a decade, many others built regional econometric models as doctoral dissertations, one becoming the blueprint for the National Regional Impact evaluation system of the US Department of Commerce. Today most states and many companies maintain such forecasting models. This method, too, has become an industry with many companies selling forecasts to private subscribers. A particular noteworthy case is the eclectic REMI model of George Treyz, which grew from a design at a regional science meeting in 1978 into a widely used commercial product.

Modeling land use and transportation flows within a metropolitan region proved to be arguably the most ambitious search for practical methods, and the most spatially explicit. Its intellectual scope is daunting: understanding the determinants of the location of industrial, commercial, residential, and other activities and of the movement of people among those locations. It combined five elements: the theoretical urban economic framework of Alonso and Lowdon Wingo, mathematical programming methods to make them operational starting with the Herbert–Stevens model of land use, gravity models of spatial interaction eventually enriched by the entropy formulations of Alan Wilson, discrete choice statistical methods eventually incorporating McFadden’s work, and the transportation forecasting framework established by Robert Mitchell and Chester Rapkin. For four decades, university research groups, notably those led by Britton Harris, Alan Wilson, and David Boyce, have produced a sustained flow of dissertations and other contributions. Yet, unlike the cases of input–output and econometric models, the pioneering research phase did not lead to widespread adoption in routine public and private practice. Dominant practice remains a set of incremental steps—predict population and employment change, decide (usually unaided by models) where those activities will locate, and then use transportation modeling software to forecast the flows among the locations. The research quest continues, large-scale operational models continue to be built (about a dozen were in use in the early 1990s), and the research leads to improvements in the transportation modeling software and other elements of the incremental approach.

2.3 Regional Development

The basic quest here is to understand differences in regional growth and development. During the early years of modern regional science it attracted the attention of two more Nobel Laureates in Economics. Gunnar Myrdal was honored in 1974 for his ‘penetrating analysis of the interdependence of economic, social and institutional phenomena.’ His work on economic theory and underdeveloped regions centered on the principle of cumulative causation, a ‘vicious circle’ of economic and noneconomic factors that led activities to cluster in certain regions and caused regional inequalities. Douglass North was honored in 1993 for ‘for having renewed research in economic history by applying economic theory and quantitative methods in order to explain economic and institutional change.’ He studied regional economic growth using exportable commodities or the export base as the key concept, pointing out the incentives regions have to reduce their processing and transportation costs to better the competitive position of their exports. An influential book by Michael Porter, The Competitive Advantage of Nations (1990), builds on these notions of clusters of activities and competitiveness of export industries. He argues that national advantage derives from successful firms concentrating geographically within a country, their competitive success stimulated by their local environment, including government policy at the regional and local level. An earlier formulation remains popular in regional policy. Based on the growth pole research initiated by Francois Perroux, governments concentrate investment in selected urban centers to stimulate the economies of their surrounding areas or draw population away from the crowded, larger metropolitan areas.

2.4 Other

Brief mention of some other work completes a broader picture of regional science. RSAI has awarded its Founder’s Medal to Martin Beckmann for research in transportation network theory, location theory, and spatial economics; William Alonso for urban economics, spatial demography, and regional development; Jean Paelinck for methods of regional analysis, spatial econometrics, and regional policy; and David Boyce for transportation network models and urban transportation planning methods. The three most cited articles in the International Regional Science Review are on population dispersal from major metropolitan regions internationally (D. Vining 1978), intra-urban residential mobility (J. Quigley and D. Weinberg 1977), and the shift of the US population to nonmetropolitan areas (C. Beale 1977). In the Journal of Regional Science they are on the distribution of residential activity in urban areas (J. Herbert and B. Stevens 1960), determinants and consequences of migration (M. Greenwood 1985), and an efficient algorithm to solve the Weber location problem (H. Kuhn and R. Kuenne 1962). In Regional Science and Urban Economics, they are on the baby boom and the housing market (N. Mankiw and D. Weil 1989), the demand for housing characteristics (J. Follain and E. Jimenez 1985), and publicly provided inputs and economic growth (T. Garcia-Mila and T. McGuire 1992). The supplementary Bibliography: includes well-cited books and recent overviews of aspects of regional science.

Bibliography:

1. Anselin L 1988 Spatial Econometrics. Kluwer Academic Publishers, Boston
2. Fujita M 1989 Urban Economic Theory. Cambridge University Press, Cambridge, UK
3. Friedmann J, Weaver C 1979 Territory and Function. University of California Press, Berkeley, CA
4. Hansen N, Higgins B, Savoie D J 1990 Regional Policy in a Changing World. Plenum Press, New York
5. Henderson J V 1985 Economic Theory and the Cities, 2nd edn. Academic Press, Orlando, FL International Regional Science Review 1995 17(3) and 18(2) (Regional Science at 40 issues)
6. Isard W, Azis I J, Drennan M P, Miller R E, Saltzman S, Thorbecke E 1998 Methods of Interregional and Regional Analysis. Ashgate Publishing, Aldershot, UK
7. Journal of the American Planning Association 1994 60(1): 3–44 (Large Scale Models symposium)
8. Journal of Regional Science 1985/24(4) (Silver Anniversary issue)
9. Krugman P 1991 Geography and Trade. MIT Press, Cambridge, MA
10. Lahr M, Miller R (eds.) 2001 Regional Science Perspectives in Economic Analysis. Elsevier, Amsterdam
11. Markusen A 1985 Profit Cycles, Oligopoly, and Regional Development. MIT Press, Cambridge, MA
12. Miller R E, Blair P D 1985 Input–Output Analysis Foundations and Extensions. Prentice-Hall, Englewood Cliffs, NJ
13. Nijkamp P (ed.) 1986 Handbook of Regional and Urban Economics. North Holland, Amsterdam
14. Plane D A, Rogerson P A 1994 The Geographical Analysis of Population. Prentice-Hall, Englewood Cliffs, NJ
15. Wilson A G 1970 Entropy in Urban and Regional Modeling. Pion Press, London