Theme/Relevant Ministry:

URBAN DEVELOPMENT/ LAND USE/ REAL ESTATE:
M/o Finance; Finance department Punjab; FBR; Punjab Excise & Taxation department

Project Brief:

This study undertook the urban immovable property valuation in Lahore and Faisalabad, using big data and advanced spatial analysis techniques to explore the significant impact of location-specific parameters on the urban immovable property prices. To compute the immovable property values, we employed Big Data analytics in Geographic Information System (GIS). The traditional hedonic price models give little importance to the spatial characteristics of individual housing units and revolve around the structural attributes of houses. However, the spatial heterogeneity should be considered while appraising the residential property prices since the house characteristics may vary over space. To address it, we established different valuation models based on the ordinary least square regression and the Fast Geographic Weighted Regression (FastGWR) model, a scalable open source implementation of python and Message Passing Interface (MPI) that can process millions of observations. These valuation models estimated the total net worth of the residential real estate market in both the study areas. The results demonstrate the excellent performance of our valuation models and display the spatial heterogeneity with higher accuracy. The valuation models explained the relationship of explanatory variables to response variable up to 75% for Faisalabad and around 85% for Lahore. Results show that the floor area, proximity of health facilities, recreational sites and market places add premium to, while nearness of educational institutions, worship places and solid waste transfer stations or dumping sites lessen the property values in both the cities.

Public Policy Relevance:

This study argues that the current system of valuation of immovable properties (DC rates and FBR rates) by the government agencies is inefficient, non-scientific, and inconsistent. Due to the poor official property valuation system and low regulatory oversight, most of the gains go unreported, which in turn gives rise to black economy practices and loss of revenue for the national exchequer. To compute immovable property values, the study uses Geographic Information System (GIS) and Big Data analytics. The study helps to establish a more sophisticated system of valuation of immovable properties based not only on the structural attributes but on the spatial variables as well.

Unedited Working Paper and Policy Brief prepared for the Second RASTA Conference can be downloaded from the link: https://pide.org.pk/rasta/2nd-rasta-conference/