Skip to content

Understanding drivers of urban growth and forecasting

The spatial pattern of economic activity in urban areas (analysed and modeled in a spatially explicit manner) is a major issue that crosscuts the fields of economic geography, spatial economics, regional studies and land use science. My research focuses on causes and dynamics of urban growth through the perspectives of land use change science and firm location choice.

I have analyzed commercial and industrial land use change decisions in suburban and exurban areas of the northern Washington D.C. metropolitan area targeting the better understanding of the widely observed phenomenon of employment decentralization (job sprawl) in North American cities. In view of the actual and potential impacts of this form of urban development, I argue for the need for a balanced attention to both residential and job decentralization due to inconclusive evidence on whether people follow jobs or firms follow people towards suburbia and exurbia. The novelty of the approach is the employment of a highly applicable spatially explicit land use change analysis framework on questions that have only marginally been addressed through the lenses of land use science; furthermore, I address the issue of solely studying drivers of residential land use change in the urban fringe but not the drivers of commercial and industrial land uses. In particular, I test the significance of market conditions, local government growth controls, such as zoning and moratoria on development, and other factors as drivers of industrial and commercial land use change through discrete choice statistical models. This paper was published in the Journal of Land Use Science and is co-authored with Jackie Geoghegan.

In a paper titled “Determinants of early spatial patterns of foreign direct investment (FDI) in the Pearl River Delta of Southeast China” I explore why foreign firms locate where they do since this decision is hypothesized to have important consequences to the trajectories of urban and localized economic growth due to path dependence effects. I examine factors influencing the location choice decision made by foreign investors in southeast China for foreign firms that began operations in the Pearl River delta during the period 1979 to 1987. I modeled the decision for location at the (sub-county) district level using discrete choice modeling techniques. I find evidence of early spatial agglomeration forces (in particular, localization economies), preferential localized policies through special economic zones and same-home foreign investor attraction forces as well as indirect empirical support for the thesis of kinship and personal contacts as drivers of early FDI in the PRD. The paper is in manuscript format and is co-authored with Karen C. Seto.

A major concern regarding urban development in rapidly growing cities of the developing world is where exactly urban growth will occur. My research has focused on computational and statistical models that predict probabilities of the location of urban growth. In a paper published in Environment and Planning B in 2007 I developed a novel urban growth model applicable to data-sparse environments. Using easily obtainable data such as land use change maps, transportation networks and administrative boundaries, the model can inform policy makers on the issue of most probable location and timing of urban growth. The model explicitly captures important aspects of the decision-making process through its flexibility for testing alternative development scenarios. It also assists in the policymaking process by addressing model uncertainty – integrating the idea that a best or true model of a system does not exist and different model specifications have to be taken into account according to appropriate weighing schemes. A statistical decision making module is also proposed in which a decision maker defines loss functions that can depend on deviations from urban growth and pattern targets. A different perspective on this work, discussing the policy relevance component in more detail, was published as a book chapter in the edited book ‘Land Use Science: Science, Policy and Management’ by Richard Aspinall and Michael Hill.