3.1.2 Evolution of Urban Land Value Research

Urban land value has been one of the central topics in urban economics, land economics, geography, and urban planning for more than two centuries. Throughout this period, research has evolved from simple location-based explanations toward increasingly sophisticated models incorporating spatial analysis, geographic information systems (GIS), big data, and artificial intelligence.

The evolution of urban land value research reflects not only advances in scientific methods but also changing perspectives on how cities function and how value is created within complex urban systems.

Although different disciplines have approached urban land value from different perspectives, they share the common objective of understanding the relationship between land, location, human activities, and economic development.

The development of urban land value research can be broadly classified into several major stages.

(1) Classical Location-Based Theories

The earliest studies emphasized the relationship between location and land value.

Land value was primarily explained by accessibility to economic centers and transportation costs.

These theories established the fundamental principle that location is one of the primary determinants of urban land value.

However, they generally assumed relatively simple urban structures and stable economic conditions.

(2) Urban Economic Models

With the development of modern urban economics, researchers introduced analytical models describing the relationship between land markets, urban growth, population distribution, and land rent.

Urban land value became closely associated with market equilibrium, economic productivity, and land-use competition.

These models greatly improved theoretical understanding but often represented cities as simplified economic systems.

(3) Spatial Analysis and GIS

The emergence of Geographic Information Systems (GIS) transformed urban land value research by enabling spatial visualization and quantitative analysis.

Researchers could now evaluate accessibility, neighborhood characteristics, infrastructure, environmental quality, and spatial heterogeneity using digital geographic data.

This stage marked a transition from purely economic analysis toward spatially explicit modeling.

(4) Data-Driven and Artificial Intelligence Approaches

Recent advances in remote sensing, big data, machine learning, and artificial intelligence have significantly improved the accuracy of land value prediction.

Modern valuation models can process large volumes of spatial, socioeconomic, and environmental data to estimate property values with unprecedented precision.

Nevertheless, most AI-based approaches remain predictive rather than explanatory.

They are highly effective at estimating land prices but generally provide limited understanding of the underlying mechanisms responsible for value formation and spatial evolution.

(5) Toward an Integrated Scientific Framework

The increasing complexity of contemporary cities has revealed the need for a broader scientific perspective.

Urban land value is now recognized as the outcome of multiple interacting processes involving infrastructure investment, accessibility, economic activities, land-use change, governance, environmental conditions, and social dynamics.

Understanding these interactions requires more than accurate prediction models; it requires a theoretical framework capable of explaining how urban value is created, propagated, transformed, and evolves over time.

This recognition provides the scientific context for the development of the Urban Land Value Framework (ULVF).

Key References

The discussions presented in this section are primarily based on the following references:

[4] Alonso, W. (1964). Location and Land Use: Toward a General Theory of Land Rent.

[5] Fujita, M. (1989). Urban Economic Theory: Land Use and City Size.

[6] Longley, P. A., Goodchild, M. F., Maguire, D. J., & Rhind, D. W. Geographic Information Systems and Science.

[7] Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning.

[8] Batty, M. (2013). The New Science of Cities.